Gastrointestinal Physiology and Results following Bariatric...

74
– the

Transcript of Gastrointestinal Physiology and Results following Bariatric...

– the

”Οί παχὲες σφὸδρα κατὰ φὺσιν, ταχυθὰνατοιγὶνονται μᾶλλον τῶν ἰσχνῶν”

“Those who are constitutionally very fat have lesspower to successfully resist a severe disease”.

Hippocrates aphorisms II

Till Anna, Ebba och Agnes

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Hedberg J, Hedenström H, Nilsson S, Sundbom M, Gustavsson

S. (2005) Role of gastric acid in stomal ulcer after gastric by-pass. Obesity Surgery, 15(10): 1375-1378

II Hedberg J, Hedenström H, Sundbom M. Wire-less pH-metry at the level of the gastrojejunostomy after Roux-en-Y gastric by-pass-a novel use of the BRAVO™-system. Submitted

III Hedberg J, Hedenström H, Karlsson FA, Edén-Engström B, Sundbom M. (2010) Gastric emptying and postprandial PYY-response after biliopancreatic diversion with duodenal switch. Obesity Surgery, E-pub ahead of print sept 2010

IV Hedbeg J, Gustavsson S, Sundbom M. Superior weight loss and glucose control three years after duodenal switch compared to Roux-en-Y gastric bypass. Submitted

Reprints were made with permission from the respective publishers.

Contents

Introduction ................................................................................................... 13

Background ................................................................................................... 14 Definitions and diagnostic methods ......................................................... 14 Epidemiology of obesity .......................................................................... 15 Risks of obesity ........................................................................................ 15 Diets and behavioral treatment of obesity ................................................ 16 Medical treatment of obesity .................................................................... 16 Surgical approaches to the treatment of obesity ....................................... 16

Gastric bypass, RYGBP ....................................................................... 17 Biliopancreatic diversion with duodenal switch, DS ........................... 19

Complications to RYGBP and DS ........................................................... 20 Acid secretion after RYGBP .................................................................... 23 Gastric emptying and pylorus function .................................................... 23 PYY .......................................................................................................... 24

Aims of this thesis ......................................................................................... 25

Patients and methods ..................................................................................... 26 Paper I ...................................................................................................... 26 Paper II ..................................................................................................... 26 Paper III .................................................................................................... 28 Paper IV ................................................................................................... 30 Ethics ........................................................................................................ 33

Results ........................................................................................................... 34 Paper I ...................................................................................................... 34 Paper II ..................................................................................................... 35 Paper III .................................................................................................... 40 Paper IV ................................................................................................... 42

General discussion ........................................................................................ 48

Conclusions ................................................................................................... 55 Paper I ...................................................................................................... 55 Paper II ..................................................................................................... 55 Paper III .................................................................................................... 55 Paper IV ................................................................................................... 55

Svensk sammanfattning (summary in Swedish) ........................................... 56 Delarbete I ........................................................................................... 56 Delarbete II .......................................................................................... 56 Delarbete III ......................................................................................... 57 Delarbete IV ........................................................................................ 57 Konklusioner ....................................................................................... 58

Acknowledgements ....................................................................................... 59

References ..................................................................................................... 62

Abbreviations

BMI Body mass index BPD Biliopancreatic diversion

(=Scopinaro procedure) CT Computed thomography DS (=BPD-DS) Biliopancreatic diversion with duo-

denal switch EBW Excess body weight E(B)WL Excess (body) weight loss JIB Jejunoileal bypass LRYGBP Laparoscopic Roux-en-Y gastric

bypass PYY Peptide YY RYGBP Roux-en-Y gastric bypass SAGB Silicone adjustable gastric banding VBG Vertical banded gastroplasty

13

Introduction

A major dietary shift took place among humanoid primates about two mil-lion years ago with increased reliance on meat1. This was made possible by tool use and social organization and the changed diet was instrumental for human evolutionary divergence from our primate relatives2. The adaptation to more energy dense food compensated for the increased metabolic cost of a larger brain size3 and the drive to obtain energy dense food is therefore an integral part of human nature. The ability to store calories when possible is important in an environment with seasonal differences in food availability4. A distinct change in energy availability occurred with the invention of agri-culture and animal husbandry about 10000 years ago5. In recent decades, food availability has increased in large parts of the world, at the same time as the requirements for manual labor has decreased, resulting in a major in-crease in obesity6-9. Today overweight is the most common form of malnutri-tion10.

Many efforts to treat obesity have been made. The use of diets of varying composition in order to loose weight have been advocated since the time of Hippocrates11. In the 20th century, medical treatment has been introduced. These drugs were introduced with great hope of solving the problem of obes-ity, but systematic evaluation has shown limited efficacy in the long run12. Bariatric surgery was introduced in the 1950s and has been refined up until today13-21. The refinement of surgical techniques has led to increasing safety and efficacy and today, there is a wide spread consensus that surgery is the best treatment modality for the morbidly obese patient22. Bariatric surgery (from the Greek word for weight: baros) is currently expanding rapidly in the world23, as are the efforts to optimize the operations and studying the post-operative physiology. In the present thesis the focus will be on the re-sults of surgical treatment for obesity in the adult patient. The topic of over-weight and obesity in children and adolescents is also very important, but beyond the scope of the present summary.

14

Background

Definitions and diagnostic methods The most commonly adopted method to define overweight and obesity is to calculate the BMI (Body Mass Index). BMI is defined as body weight in kg divided by body length in meters squared. This gives the unit measure of kg/m2, which is often left out. With this method a normal BMI is 18.5-24.9. Obesity is divided into classes of severity; Class I is BMI 30-34.9, Class II is BMI 35-39.9 and class III is BMI of 40 or above, also often referred to as morbid obesity 24, 25. With increasing numbers of obese patients with very high BMI, further subdivision has been developed, and patients with BMI>50 are referred to as super-obese, and those with BMI>60 as super-super obese. Many other definitions of overweight and obesity have been in use over the years. The Metropolitan Life Insurance Company has defined the weight at which the mortality is the lowest for any given body length26. In the American literature Excess Body Weight (EBW) is often used. This is the weight above the ideal weight according to these tables. Excess weight loss (EWL) or excess body weight loss (EBWL) is often used as a measure of success of weight reduction programs. It is often expressed as % EBWL, which refers to the weight reduction divided by the former EBW.

Although BMI is the most commonly used description of weight, it has some drawbacks. It does not reflect the body composition, and for instance, someone with a very big muscular mass can have a high BMI without health risks. Nor does it take into account the distribution of fat in the body. Vis-ceral fat is known to be associated with more cardiovascular risks than sub-cutaneous fat27 and waist circumference as well as waist to hip ratio ad-dresses this. To refine measurements of body composition, more advanced diagnostic methods, such as bioimpedance28, air displacement pletysmo-graghy (the so called BOD POD)29 or radiological methods30 have been de-veloped but are generally not used outside scientific studies. BMI is a very simple and straightforward measurement and it can also be calculated with self-reported weight. This gives an opportunity to evaluate treatments with-out a visit to the health care facility and a rising use of BMI is seen in the literature.

15

Epidemiology of obesity In the last 25 years, obesity has become a worldwide epidemic, beginning in the western countries. The prevalence has been estimated to be as high as 1.7 billion people10, and for the first time in the western civilization the coming generation may have a shorter life expectancy than its parents as a result of this31. In the US, more than one third of the population is obese and nearly two thirds overweight8. This represents more than a 100% increase of the obese population in the last 25 years. Given that other western countries often tend to follow USA, it is encouraging to see that since 2004, the trend of increasing obesity may have abated in the US32. In Sweden, obesity is rising fast and it is mostly widespread in the lower socioeconomic popula-tion33, 34. In the male population, 52 % have a BMI>25, of which 11 % are obese. For the female population these percentages are 36 and 9 respectively 35.

Risks of obesity Mortality in cardiovascular diseases, such as stroke, coronary heart disease, congestive heart failure and cardiomyopathy, is higher in the overweight and the obese36, 37. In addition to mortality related to these risk factors, mortality is also independently linked to overweight per se38-40. A man in his twenties with a BMI>45 statistically loses 13 years of his life compared to a normal weight person, and this can represent a loss of 22% of expected life span41.

Hyperlipidemia and hypertension are linked to obesity42. Moreover, the visceral fat accumulated in morbidly obese patients gives rise to an inflam-matory drive, which is an important factor in the development of artheroscle-rosis36, 43.

Type 2 diabetes mellitus has an independent relationship with a high BMI, as well as with weight gain in adult life44, 45. The level of insuline resis-tance and risk of diabetes is also correlated to the amount of visceral fat46.

Obstructive sleep apnoea, a condition with nighttime desaturations due to apnoeic episodes, is another important comorbidity in obese patients. Of the patients with sleep apnoea, 70% are obese47. This condition leads to daytime drowsiness and risk of accidents, but is also in itself a risk factor for cardi-ovascular disease47, 48.

Increasing evidence for the relationship between obesity and cancer are mounting. Several types of cancer are linked to obesity, abdominal fatness or both. These are adenocarcinoma of the esophagus, pancreatic cancer, colo-rectal cancer, breast cancer in postmenopausal women, endometrial cancer, kidney cancer and gallbladder carcinoma49. It is estimated that the proportion of cancer attributable to overweight or obesity in the US is as high as 14 % in men and 20 % in women50.

16

Diets and behavioral treatment of obesity A plethora of self-help literature is on the market. The claim to have found the solution to overweight is a good base for commercial success. Weight-loss in the short term is not hard to accomplish, but weight maintenance over time is very hard to achieve. There is support in the literature that weight maintenance is possible for up to 20% of patients undergoing diets, when defined as loosing 10% of their original body weight for at least a year51. In the few studies with long-term follow up these figures are less encourag-ing52.

Medical treatment of obesity Over the years, many drugs have been tried in the pharmacological treatment of obesity. Different amphetamines have been in use since the 1940s53. To-day, there are two drugs for treatment of obesity on the Swedish market: • Orlistat is a lipase inhibitor and decreases the ability of pancreatic lipase

to absorb intestinal fat with about 30%. It has side effects related to this mechanism of action in the form of steatorrhea. The weight loss has been reported to be 8-10 kg after one year and 5-8 kg after two years12,

54. • Sibutramin is an inhibitor of serotonine and noradrenaline reuptake in

the central nervous system. This leads to decreased feeling of hunger, and a decreased calorie intake. It is contraindicated in patients with co-ronary heart disease, congestive heart failure or a history of stroke. The weight effect is similar to Orlistat and positive effects on serum lipids have been shown12, 55.

Surgical approaches to the treatment of obesity The only treatment with proven effect on the negative consequences of mor-bid obesity is surgical modification of the gastrointestinal tract56. Over the years, various surgical principles have been utilized with the goal to achieve lasting weight-loss whilst minimizing undesirable side effects.

The jejunoileal bypass or shunt (JIB), in which a large part of the small intestine is bypassed was first described in the 1960s57. Initial encouraging results led to a rise in popularity of this method. However, metabolical me-dium to long-term complications in the form of liver imparement, renal stones and electrolyte disturbances as well as problems with weight regain eventually led to abandonment of this procedure13, 58-61.

In the 70’s, horizontal gastroplasty was used in an effort to hinder exces-sive food intake without interfering with the normal food pathway14. This

17

technically straightforward operation was well tolerated but had the problem of dilatation of the fundus and staple line disruption, both of which led to weight regain.

The vertical stapled gastroplasty (VBG) was developed in the 80’s, in an effort to overcome these problems 15. In this operation a fixed band is placed from a hole made in the corpus of the stomach to the lesser curvature. From this hole a staple line is constructed up to the angle if His and in this fashion all ingested food is led into a small pouch. When constructed along the lesser curvature, the pouch does not dilate and the synthetic band maintains a tight outlet. Band complications, such as band erosion, esophagitis and dysphagia as well as weight regain due to staple line disruption or adaptation of oral intake are associated with this operation62, 63.

The gastric banding procedure, where a band is placed from the pars flac-cida to the angle of His was introduced in the same decade as the VBG and it won some popularity16. Refinement of this operation resulted in the silicone adjustable gastric banding procedure (SAGB), in which an inflatable band is placed around the cardia and connected to a subcutaneous port which can be inflated or deflated in order to adjust the stoma size. As in the previously mentioned restrictive procedures, problems of esophagitis, vomiting, dys-phagia and band erosion in combination with suboptimal long-term weight results are all described in the litterature64-66. Despite this, the SAGB opera-tion is gaining in popularity in the US whereas it is less commonly used in Europe23.

Gastric bypass, RYGBP Gastric bypass for treatment of obesity was fist described by Ed Mason in 196717. In this first series, the proximal pouch was constructed by dividing the fundus horizontally and a loop gastrojejunostomy was added, thus creat-ing an operation similar to the Billiroth II procedure, but without the distal resection. A number of modifications of the technique have been made over the years of which the most important ones are the construction of a Roux-limb instead of a loop67, and making the proximal pouch along the lesser curvature instead of the fundus68. These alterations, resulting in the Roux-en-Y gastric bypass (RYGBP), have minimized the problem of bile reflux from an afferent limb69, and pouch dilatation in the fundic area of the proximal pouch (Figure 1).

With the introduction of surgical staplers, the level of technical complex-ity decreased and the use of gastric bypass became more widespread even though restrictive operations dominated the bariatric field throughout the 80’s. When the laparoscopic era started in the early 90’s, the increased popu-

18

larity of simpler restrictive procedures led to smaller numbers of gastric by-pass operations being performed. In 1994, Wittgrove published the first se-ries of five laparoscopic Roux en Y Gastric Bypass (LRYGBP) operations with encouraging results18. Handport-assisted laparoscopic gastric bypass does not show any advantages to open surgery in a randomized trial70.The totally laparoscopic approach however, resulted in fewer wound complica-tions, shorter hospital stay, less morphine use, less incisional hernias and a faster rise in quality of life71-73.

Today, the number of patients undergoing gastric bypass is increasing ra-pidly. In Sweden, solely restrictive procedures have become rare and LRYGBP has come to dominate the field. As is the case with the before mentioned bariatric operations, there are specific complications with this technique as well, and they will be discussed in more detail below.

Figure 1 Roux-en-Y gastric bypass, RYGBP. A small proximal gastric pouch is created below the gastroesophageal junction from the lesser curvature. The Roux-limb is created by dividing the jejunum distal to the ligament of Treitz’ and connect-ing the aboral part to the gastric pouch. Continuity of bile and pancreatic flow is restored with an enteroenterostomy

19

Biliopancreatic diversion with duodenal switch, DS The duodenal switch is a combination of two operations. De Meester origi-nally described the duodenal switch for the treatment of duodenogastric ref-lux74. The biliopancreatic diversion (BPD) was first described in 1979 by Scopinaro21 and is also known as the Scopinaro procedure. It is devised to induce a decreased fat absorption by creating a distal Roux-en-Y and thus only allowing uptake of fat in the distal ileum, where the bile and pancreatic juices meet the food-stream. In addition to this, the Scopinaro procedure includes a distal gastrectomy to reduce the gastric volume. Hess later used the Scopinaro procedure after failed restrictive procedures, but the distal gastrectomy proved difficult in revisional operations due to adhesions, and the relatively large gastric remnant gave rise to stomal ulcers in the Roux limb. He therefore created a tube along the lesser curve of the stomach, ie. sleeve gastrectomy, and added the duodenal switch. Hess and Marceau pub-lished their first series of biliopancreatic diversion with duodenal switch (DS) in 199819, 20. The promising results from revisional procedures led them to adopting DS as a primary bariatric procedure. Hess et al., with more than 1400 operations and follow up of 92% after ten years75, have proved the method to be safe and effective. DS has not, however, become widely popu-lar and we believe that this is partly due to the technical difficulty of the operation, especially in the laparoscopic setting, and partly due to concern of metabolical and/or gastrointestinal disturbances following the operation. Comparisons between RYGBP and DS show superior weight loss after DS76-

78 with one exeption79.

20

Figure 2 Biliopancreatic diversion with duodenal switch, DS. The small intestine is divided 250 cm oral to the ileocaecal valve and the cephalad part is connected to the distal ileum one meter from the ileoceacal valve, thus creating the one meter long common channel. A sleeve gastrectomy is then performed and the duodenal bulb divided two to four centimeters distal to the pylorus. Finally, a duodenoileostomy is created to restore continuity of the gastrointestinal tract

Complications to RYGBP and DS

Leakage The most feared complication in the immediate postoperative period is a leakage in one of the anastomoses. It is a life-threatening complication, which demands prompt intervention with repair (if made possible by early detection), adequate drainage and broad-spectrum antibiotics80, 81. Symptoms include disproportional epigastric pain, tachycardia, shortness of breath, anuria and hypotension. Leakages are more frequent after RYGBP than after DS20, 75, 80, 82.

21

Bleeding Postoperative bleeding can arise after all types of surgery, and bariatric sur-gery is no exception. In addition to bleeding into the peritoneal cavity, intra-luminal bleeding in the gastrojejunostomy or other anastomoses can arise in the early postoperative period, and require reoperation83, 84.

Pulmonary Embolism Although being a rare complication, pulmonary embolism is one of the lead-ing causes of death after bariatric surgery85, 86.

Small bowel obstruction With the intestinal transpositions involved in modern bariatric operations, mesentery defects are constructed. These can become sites of internal hernia-tion. Intermittent colicky pain after meals should raise suspicion of this complication. Diagnosis can be difficult with ordinary computed tomogra-phy (CT), since the affected small bowel can be on the biliopancreatic limb, and thus not cause classical radiologic signs of obstruction, since it is not in the food-stream. It is important that the surgeon reviews the radiographic pictures. CT can be deemed normal by the radiologists in 60% of cases with internal herniation when the pathology is clear to the operating surgeon87. However, coeliotomy sometimes has to be undertaken on clinical suspicion alone since an overlooked gangrene can be disastrous for the patient. Post-operative adhesions can also lead to obstruction82, 88.

Nutritional deficiencies A number of nutritional deficiencies are described after bariatric surgery. Protein malnutrition are described after RYGBP89 and DS75, 90. Folate defi-ciencies can occur and vitamin B12 is not readily absorbed since the produc-tion of intrinsic factor is decreased91. Iron and calcium are normally ab-sorbed in the duodenum and proximal jejunum, both of which are bypassed in modern bariatric surgery, why supplementation of these are necessary92. Fat-soluble vitamins; A, D and K are also reported to decrease in serum over time after DS93. Thorough preoperative information on the importance of maintainance to nutritional supplementation programs as well as careful follow-up of these patients is therefore important.

22

Stomal ulcer

Figure 3 Stomal ulcer distal to the gastrojejunostomy

Stomal ulcer (Figure 3), also called marginal ulcer is a well-known clinical entity after gastric resections94, 95. This clinical problem also occur in <1%-9% after RYGBP96-99, often requiring long-term medication and sometimes reoperation. It can cause significant bleeding, leading to the need of transfu-sions and endoscopic intervention, or even perforation with peritonitis100, 101.

Symptoms of stomal ulcer typically arise during the first month after sur-gery102, and consist of dull epigastric pain and/or nausea which are often aggravated by food intake. Stenosis with dysphagia can be the only symptom as can upper gastrointestinal bleeding with anemia or signs of hypovolemia. The diagnosis is readily made with endoscopic examination and treatment with proton pump inhibitors (PPI) often suffice. A higher dose than normal is often required and relapse of symptoms after cessation of the medication is not uncommon. Sometimes life-long therapy is warranted.

The observation that stomal ulcers often occur in patients with gastro-gastric fistulae from the bypassed stomach led to the assumption that acid from the bypassed stomach caused these ulcers103, 104. Efforts have been made to reduce the size of the pouch, in order to minimize the amount of parietal cells above the gastrojejunostomy98. The practice of complete trans-section of the proximal pouch reduces the number of gastro-gastric fistulae.

23

However, the problem with stomal ulcers does not resolve completely with these measures even if the frequency seems to go down98.

Death Mortality rates after bariatric surgery varies in the literature between 0.1 and 7,6 %105, 106. The bariatric postoperative mortality in Sweden is 0.16%107. Although this figure is low, every postoperative death is a tragedy for all concerned in this type of preventive surgery and this risk must always be communicated to the patient and preferably next of kin before a decision to operate can be made.

Acid secretion after RYGBP The parietal cells of the gastric mucosa produce hydrochloric acid by active-ly transporting H+ -ions into the gastric lumen with H+-K+-ATPase. Cloride ions are transported passively into the lumen. The parietal cells are activated during a meal, and the resting H+ secretion of 2 mmol/h can increase tenfold. The most important stimulant of gastric acid secretion is gastrin. This hor-mone, released by G-cells in response to vagal nerve signals, or mechanical stimulation of the gastric mucosa, induces histamine secretion from ECL-cells which in turn acts in a paracrine way to stimulate H+-K+-ATPase in the parietal cells108, 109. After gastric bypass, acid production in the bypassed stomach is suppressed110 and measurements with a perfused orogastric tube have failed to demonstrate significant amounts of acid in the proximal pouch after RYGBP111. This group has demonstrated that the acid-secreting parietal cells are present in the cardiac area all the way to the gastroesophageal junc-tion112, and one of the aims of this thesis was to evaluate the pathogenic role of acid production from these cells in ulcer formation at the gastrojejunost-omy after RYGBP.

Gastric emptying and pylorus function Gastric emptying is normally characterized by two phases. The first phase is called lag-phase. It can be measured as the time from ingestion to the first appearance of solids in the proximal small bowel. A linear emptying-phase then ensues113, 114. The pylorus acts as a regulator of the flow from the sto-mach to the duodenum and vice versa. The pylorus is innervated mainly from the vagal nerve115. It has a number of responses to different stimuli. For instance, when the duodenum is exposed to fat or acid, the pylorus responds with increased baseline activity and phasic contractions116. Unlike most sphincters of the gastrointestinal canal, the pylorus maintains a lumen most of the time117.

24

PYY A number of gastrointestinal peptides are involved in gastric emptying and satiety regulation118, 119. One of these, peptide YY (PYY), is studied in this thesis. PYY is a 36-amino-acid peptide of the neuropeptide Y family. It is secreted from the enteroendocrine L-cells in the small and large intestine with increasing concentration distally120. The active N-terminally truncated PYY3-36 inhibit gastric emptying and increase intestinal transit time121, 122. The ileal brake, which slows transit through the gastrointestinal tract in re-sponse to fat, protein and carbohydrates in the ileum, depend on PYY among other transmittors123. Active PYY3-36 has high affinity to Y2-receptors in the hypothalamus, inducing satiety in humans124 and non-operated obese sub-jects have low124 or normal125-127 levels of PYY. While its effect on food intake in humans has been disputed128, elevated levels of this PYY is seen after JIB129. A recent study shows decreased food intake after orally adminis-tered PYY130 and it could be one among other mechanisms whereby food intake is decreased after bariatric surgery.

25

Aims of this thesis

In light of the massive rise in bariatric surgery in Sweden and international-ly, the need for study of the postoperative physiology increases. Certain complications of bariatric surgery will increase in prevalence and efforts to understand the underlying patophysiology is vital for development of correct treatment strategies. The aim of this thesis was to investigate certain changes in gut physiology, and results of bariatric surgery. The specific aims were:

I to study the pathogenetic role of hydrochloric acid in the formation of stomal ulcer after gastric bypass with a case-control study of pH in the proximal pouch in patients with stomal ulcer. A secondary aim was to describe results of revisional operations for stomal ulcer.

II to introduce a new use of wire-less pH-monitoring and to describe the level of the acidity at the gastroenteroanastomosis after RYGBP in asymptomatic patients.

III to describe gastric emptying after DS and to study PYY levels after DS in relation to age-matched control subjects. To investigate the occur-rence of dumping symptoms after DS.

IV to compare DS and RYGBP in a randomized controlled trial. Major out-come variables were perioperative complications and weight results. Ga-strointestinal symptoms and biochemical evaluation more than two years after surgery were secondary outcome variables.

26

Patients and methods

Paper I Six patients (one male) with endoscopically verified stomal ulcers were in-cluded. The time from operation to diagnosis was in median 3.5 weeks (3 weeks to 5 years). Barium swallow x-ray studies were obtained to exclude gastro-gastric fistula. Control subjects were six asymptomatic women who had undergone RYGBP at least 5 years earlier. After discontinuation of pro-ton pump inhibitor use for at least one week, a pH-sensitive probe (Digitrap-per.pH, Medilab) was passed transnasally to the proximal pouch as guided by calculated distance and pH-response. The probe was left in place for four hours during which time the subjects could eat and drink at their leisure. The proportion of time with pH<4 was calculated and compared between stomal ulcer patients and controls by Wilcoxon´s Rank-sum analysis. Three revi-sional operations were performed. Two of these included total resection of the gastroenterostomy as well as reduction of the proximal pouch. In the third operation, a large part of the proximal pouch could be resected without touching the gastroenterostomy.

Paper II

Patients In total, 25 patients (4 men), median age 44 years (range 31-60) and BMI 29.3 (range 22.7-40.1) kg/m2, who had undergone RYGBP 4 years (range 2-8 years) earlier were studied. Twenty-one of these patients had no symptoms or history of stomal ulcer, nor did they use PPI or NSAID. Four patients with symptoms from the epigastric area and who were under PPI-treatment (con-tinous or occasional) were also recruited. Their data are reported separately. In an attempt to evaluate reproducibility, two of the 21 asymtomatic patients underwent double measurements. In one patient the measurement was re-peated after one month and in the other patient two Bravo-capsules were applied side by side.

All operations had been performed at the Department of Surgery, Uppsala University Hospital. A small proximal pouch was constructed at the lesser curvature by dividing the stomach horizontally 5 cm below the gastroeso-

27

phageal junction with a cutting linear stapler (45-mm blue cartridge). The pouch was then totally transected using one to three vertical firings with identical cartridges. After placing a purse sting suture just above the initial horizontal staple line, it was cut and the anvil of a 25-mm circular stapler was entered into the pouch. This and the forthcoming stapling of the upper anastomosis reduced the length of the pouch by about one cm. The jejunum was divided 30 centimeters from the ligament of Treitz and a classical Roux-limb of 70 centimeters was constructed. Finally, the gastrojejunostomy was performed end-to-end with the circular 25-mm stapler. For study proposes, the number of 45-mm cartridges that were used to construct the vertical part of the pouch was noted, as a rough measurement of the pouch size.

pH-metry After a fasting period of at least 6 hours, an upper endoscopy was performed. The anastomotic area was checked for ulcers and other pathology, such as gastrogastric fistula, and the distance to the incisors was measured. A cali-brated BRAVO™-capsule was then inserted transorally and followed by the endoscope, so the capsule could be applied to the anastomotic region under direct visual control (Figure 7). After applying suction (510 mm Hg) for initially one minute, later two minutes, to obtain a portion of mucosa in the positioning hole of the capsule, the locking pin was activated to secure the capsule to the wall. No fluoroscopy was used. Nine patients had lidocain spray in the pharynx, and four patients also had a mild sedation with 5 mg of diazepam intravenously. During the 48-hour registration, the study patients could eat and drink at their leisure. Data were recorded in the receiving de-vice, carried in a band over the shoulder or kept within one meter from the body.

Data analysis The pH-plots were analyzed using the software Polygram NET ® version 4.2 (Given Imaging, Yoqneam, Israel). The plots were truncated if the pattern suddenly changed and showed a consistently neutral environment indicating a premature capsule detachment (Figure 8). The total registration time before detachment, time with pH<4, number of episodes with pH<4 as well as number of episodes with pH<4 for more than 5 minutes were registered. The percentage of time with pH<4 was calculated, as well as its distribution dur-ing daytime (08.00-22.00) and night (22.00-08-00). The graphs were adapted for publication with Pixelmator version 1.4.1, Pixelmator team ltd, London, UK.

28

Statistics Descriptive statistics were used. In calculating relationship between the number of cartridges needed for the vertical part of the pouch and level of acidity, linear regression was used and when comparing the proportions of daytime and night, Wicoxons matched pair test was used. p<0.05 was consi-dered significant. In the two patients who had double measurements, the plot with longest registration time was included in the main study.

Paper III

Patients and operation Twenty patients who had undergone DS at our centre at least 18 months previously (median 3.5 years) volunteered to participate in a scintigraphic study of gastric emptying. The group, which included 12 men, had a median age of 43 years (range 31-57) and a median BMI of 31.1 kg/m2 (range 21.8-46.3). They were weight stable and had had a BMI before surgery of 51.7 kg/m2. Out of these 20 patients, 10 patients also volunteered for PYY meas-urements after a standardised meal (5 men, median age 38 years and median BMI 27.0). For comparative reasons, the PYY-results were related to post-prandial PYY-measurements of 9 non-operated, normal-weight, age-matched subjects (median BMI 22.4 kg/m2, range 19.9-25.8) from previ-ously published work 131.

The DS had been performed at the Department of Surgery at Uppsala University Hospital. The operation consisted of transection of the small in-testine 250 cm oral to the ileocaecal valve. The distal end of the cephalad part was anastomosed to the ileum 100 cm from the ileocaecal valve, creat-ing an alimentary limb of 150 cm and a common channel of 100 cm. The remaining small bowel, the biliopancreatic limb, was in median 3.6 m (range 2.5-5.5m). A sleeve gastrectomy was performed along the lesser curvature with cutting staplers, lateral to a 38 French bougie and a one-centimetre free margin. The resection started at the greater curvature, 5 cm oral to the pylo-rus after division of the right gastroepiploic artery. For technical reasons, the right gastric artery was divided as well as all tissue lateral to the pyloric re-gion. The duodenal bulb was transected two to four centimetres distal to the pylorus and anastomosed to the oral end of the alimentary limb with a hand-sown end-to-side anastomosis (Figure 2).

29

Gastric emptying studies The scintigraphic measurements were performed at the Department of Nu-clear Medicine after an overnight fast. A 10 MBq 99mTc-labeled omelette (1/2 egg, one table spoon wheat flour, 25 ml milk, 1-2 g margarine and salt, 77.5 kcal) and an unlabelled soft drink (150 ml, 70 kcal) were consumed within 2 minutes. The energy composition of the meal was 30 % fat, 12 % protein and 58 % carbohydrates. The solid component was reduced to a quarter of normal size used in our lab132, in order to ensure that the subjects could ingest the whole meal within two minutes. Anterioposterior scans (In-finia™ Hawkeye®4, GE Healthcare, Waukesha, Wisconsin, USA) with the subject in sitting position were then obtained immediately and repeated every 5 minutes for the first 50 minutes, then every 10 minutes up until 120 minutes in total. The region of interest was outlined in the first scan and included the whole gastric tube down to the level of the pylorus. Half-emptying time was calculated, as well as percentage of tracer remaining in the gastric tube at the above-mentioned time-points. Any lag-phase was also registered.

PYY-analyses For PYY analyses, the ten DS-patients were presented at the Clinic for Obe-sity Care after an overnight fast, which was sustained until a standardised meal was served at 1300 hours. The meal consisted of 574 kcal (21% fat, 18% protein and 61% carbohydrates). Blood samples were drawn at 0800, 1000, 1200, 1300, 1330, 1400, 1430, 1500 and 1600 hours and then stored in -70 °C prior to laboratory analyses. Serum concentrations of total PYY were measured with a commercial radioimmunoassay kit (Linco Research Inc. Michigan, USA). The kit uses radioactive iodine-labelled PYY, guinea pig PYY antiserum and goat anti-guinea pig antibody/PEG technique. Sensitiv-ity of the assay was 10 pg ml -1, specificity was 100 % for human PYY1-36, PYY3-36, [Pro34] PYY and [Leu31,Pro34] PYY and <0,1% for rat/porcine PYY1-36, PYY3-36 and neuropeptid Y. Double samples were analyzed from all patients and a mean from these were calculated. All samples were assayed in the same batch. Inter- and intra-assays coefficient of variance (CVs) were <9%.

Evaluation of dumping symptoms All patients undergoing bariatric surgery at our institution are given educa-tion concerning recommended eating behaviour postoperatively and get in-formation about both early and late dumping. Dumping symptoms were evaluated on an ordinal scale (symptoms occurring daily, weekly, monthly, yearly or never). No distinction between early and late dumping was made.

30

Statistics Descriptive statistics were used in the gastric emptying study. In the PYY analyses, repeated measures ANOVA was used and p<0.05 was considered significant. Mean and standard deviation are presented, unless otherwise stated.

Paper IV

Patients During 2004-2007, 99 patients referred to us with a BMI>48 were assessed for eligibility for the study. Nine patients were excluded on medical grounds or due to language difficulties. Of the remaining 90 patients, 47 patients (23 men, age 39.1±9.9 yrs, BMI 54.5±6) agreed to have their operation (DS or RYGBP) randomized. The remaining 43 patients (8 men, age 36.4±8.3 yrs, BMI 53.3±4.8) agreed to take part in the extended follow-up but not in ran-domization. These patients were thus free to choose type of operation (Figure 4, Table 1).

Figure 4 Inclusion flow illustrated

31

Table 1 Antropometrics and comorbidities of patients included in the study. † Stu-dents t-test, ¶ Fisher exact test

Randomized patients (n=47)

Procedure of patients choice (n=43)

Total (n=90)

DS

(n=24) RYGBP (n=23)

DS

(n=21) RYGBP (n=22)

DS

(n=45) RYGBP (n=45)

Age (years) 40.2±9.5 37.9±10.4 p=0.432† 35.8±7.3 37.0±9.5 p=0.649† 38.1±8.8 37.4±9.9 p=0.724†

Gender (M/F) 12/12 13/10 p=0.438¶ 4/17 4/18 p=0.624¶ 16/29 17/28 p=0.500¶

BMI (kg/m2) 54.5±6.7 54.5±5.6 p=0.963† 53.8±4.6 52.7±5.2 p=0.501† 54.2±5.8 53.6±5.4 p=0.637†

Hypertension 6 7 p=0.464¶ 7 8 p=0.545¶ 13 15 p=0.410¶

Hyperlipidemia 0 0 - 5 3 p=0.322¶ 5 3 p=0.357¶

Sleep apnoea 4 3 p=0.525¶ 4 0 p=0.049¶ 8 3 p=0.098¶

Diabetes mellitus

(oral/insuline-treated)

6/1 1/0 p=0.055/ p=0.511¶

5/1 2/1 p=0.187/ p=0.744¶

11/2 3/1 p=0.039/ p=0.500¶

Pre-operative evaluation All patients referred to our centre for bariatric surgery were first thoroughly examined at the centre for metabolic medicine. This included individual evaluation of internist, dietician and psychologist as well as education on postoperative diet. Abnormalities were treated before surgery.

Operations All operations were performed by laparotomy with an upper midline inci-sion, typically sparing five to ten centimetres above the umbilicus. For the DS, the same technique as in paper III was used (Figure 2). For RYGBP, the technique described in paper II was used (Figure 1), with the modification that the biliopancreatic limb was made 50 cm long (instead of 30 cm) and the alimentary limb 120 cm (instead of 70 cm). All patients received throm-boprophylaxis with high-dose LWMH and single-dose of broad-spectrum antibiotics (Cefuroxime 1.5 grams IV and Metronidazol 1.5 grams per oral-ly). In two patients in the DS-group, with symptomatic cholecystolithiasis, a cholecystectomy was also performed.

Perioperative registration Perioperative data was prospectively registered for all randomized patients, with special emphasis on postoperative problems, C-reactive protein, mor-

32

phine consumption (PCA-pump, patient controlled analgesia), time to inges-tion of fluids and solids as well as all early complications. For non-randomized patients, the data was collected retrospectively.

Follow up Postoperatively, all patients were seen at our out-patient clinic, where multi-vitamin and vitamin B12 substitution was started. Patients were then followed by dietician at 3, 6 and 12 months and internist at 6 and 12 months, at the centre of metabolic medicine. Routine laboratory examinations were also undertaken. At one year, the majority of the patients were doing fine and could be referred to their primary care physician for subsequent annual fol-low up, including treatment of comorbidities, laboratory tests and further prescription of life long nutritional supplement therapy.

Laboratory evaluation Routine chemistry, including haemoglobin, vitamin B12, folate, albumin, glucose, HbA1c, HDL, LDL and triglycerides was analysed preoperatively, at one year and three years postoperatively. The latter laboratory data was taken from primary care physician follow-up (2.9 ± 0.5 yrs postoperatively).

Study data for the extended follow-up For the purpose of this study, a mail questionnaire was sent out to all patients at least two years postoperatively (n=90). We used our standard question-naire for long term follow up; current weight, medication, subsequent sur-gery as well as an extended inquiry on abdominal symptoms (appendix 1, Paper IV). The median time to this follow up was 3.5 years (2.0-5.3 years).

Statistics Power analysis was based on the historical standard deviation in BMI-reduction in our super-obese patients. To find a difference in weight loss of 5 BMI-units with a power of greater than 90 % (two-sided t-test p<0.05), a sample size of 40 patients in each group was needed. The randomization was stratified to sex and BMI (above or below BMI 53). Normally distributed values are reported with means (standard deviation), and non-normally dis-tributed as median (range). For comparison of proportions between groups Fisher exact test was used. For continuous variables, Students t-test was employed for normally distributed data and Mann-Whitney-U test for non-normally distributed data. Mann-Whitney-U test was used for data on ordinal scales. ANOVA repeated measures analysis was used for the laboratory fol-low up. Two-sided p<0.05 was considered significant. The data have been analyzed both separately for the randomized and non-randomized groups, as well as for the total sample of 90 patients.

33

Ethics

All four studies were approved by the Regional Ethics Committee at the University of Uppsala, Sweden, and informed consent was obtained from all patients.

34

Results

Paper I No gastrogastric fistulae were found on barium swallow x-ray studies. pH measurements were feasible in all patients and no adverse events occurred. The proportion of time with pH<4 was significantly grater in the patients with stomal ulcer (median 95.5%, range 49.6-100%) as compared to the asymptomatic control subjects (median 20%, range 6.7-43.9%), p<0.01, Wilcoxons rank sum test. In fact none of the controls reached the lowest value of the ulcer patients (Figure 5).

Figure 5 Representation of proportion of time with pH<4. The percentage is shown on the y-axis and each bar represents one patient

Three of the patients underwent revisional surgery because of epigastric pain that did not subside with PPI-medication. In one, a repeated pH-measurement was made and the percentege of time with pH<4 was de-creased from 100% to 6%. This patient became symptom free. The second patient had clear reduction of her symptoms and no longer needed PPI, but she refused repeated pH-metry. In the third revisional patient, a bleeding stomal ulcer occurred while she was waiting for a new pH-measurement and she was put on life-long PPI.

35

Figure 6 pH-plots of one stomal ulcer patient (upper panel) and an asymptomatic control patient (lower panel). pH is given on the y-axis and pH 4 is marked with a horizontal line. Time of day is given in the x-axis

Paper II The BRAVO™-capsule could be positioned at the level of the gastro-jejunostomy under direct vision of the endoscope in all 25 patients (Figure 7).

Figure 7 The BRAVO™-capsule is attached to the mucosa of the gastric pouch just above the gastrojejunostomy. The pH-sensor is located in the proximal Roux-limb. The applicator is seen to the left. Squamous epithelium of the esophagus is seen in the right hand side of the picture

In two patients, it was not secured enough to the mucosa and was lost into the Roux limb when the applicator was removed. Three additional patients had early detachment, judged by the consistent neutral environment until the

36

signal was lost. These five patients were therefore excluded, giving pH data from 16 asymptomatic individuals and 4 symptomatic subjects. No gastro-gastric fistula was found.

Interpretable pH-data from the region of interest was obtained for a me-dian of 25.7 hours (6.1-47.4, n=20). Registration time increased during the study period, i.e. the time before capsule detachment (indicated by a sudden and continuous shift to neutral pH, Figure 8), since four of the five imme-diate detachments occurred in the first half of the series. In the second half, the locking pin was always attached to gastric mucosa, but leaving the pH-electrode in the jejunum, and the suction time was increased to 2 minutes. These measures yielded fewer failures and a trend towards longer registra-tion times, 26.9 vs. 18.8 hours (p=0.11).

Figure 8 Two pH curves showing; A: A successful two-day recording B: Capsule detachment after 30.6 hours (at 20.00) and subsequent loss of signal when the cap-sule passes (at 07.00)

When analyzing the 16 asymptomatic patients, the median percentage of time with pH<4 was 10.5 (0.3- 37.7) (Figure 9). When dividing the registra-tion time in day (08.00-22.00) and night (22.00-06.00), the percentage of time with pH<4 was 8.4 and 6.3, respectively, (p=0.08). The median number of episodes per hour with pH<4 was 10.2 (0.1-26.2, Table 2). No correlation was found between the number of cartridges used for the vertical division of the gastric pouch, in median 2 (1-3), and the measured level of acidity.

37

The two patients examined with two capsules had consistent results. The patient with a repeated measurement after one month had pH<4 in 0.9 vs. 0.3% of the total registration time (20.0 and 23.9 h, respectively). In the pa-tient with two capsules side-by-side, one capsule was lost after 8 hours, but during that time pH<4 was registered in 13.9% and 11.6%, respectively. The remaining capsule (11.6%), stayed in place for 21.7 hours, and during that time pH<4 was noted in 9.6% of the registration time.

Figure 9 Proportion of time with pH<4 is presented on the y-axis. Each column represents one asymptomatic patient

Tab

le 2

Dat

a of

16

cons

ecut

ive

reco

rdin

gs o

f pH

in th

e pr

oxim

al R

oux-

limb

in a

sym

ptom

atic

pos

t RY

GB

P-p

atie

nts

Pat

ien

t m

=m

ale

BM

I R

egis

trat

ion

T

ime

(h)

Ep

isod

es p

H<

4 p

H<

4 >

5 m

in

Lon

gest

E

pis

ode

pH

<4(

min

) T

ime

pH

<4(

min

) E

pis

odes

p

H<

4/h

our

% t

ime

pH

<4

Day

%

tim

e p

H<

4 N

igh

t T

otal

% t

ime

pH

<4

1 37

.2

34.1

22

8 10

21

21

7 6.

7 8.

4 17

.8

10.6

2 35

.1

7.8

204

1 10

85

26

.2

- -

18.1

3 34

.4

6.1

99

6 13

95

16

.2

- -

26.0

4 30

.0

11.1

23

2 1

8 12

5 20

.9

- -

18.8

5 27

.3

22.2

7

0 2

3 0.

3 0.

3 0.

2 0.

4

6(m

) 40

.1

27.5

55

4 34

19

62

2 20

.1

39.1

34

.0

37.7

7 26

.0

20.8

36

1 15

24

31

3 17

.4

32.4

6.

3 25

.1

8 24

.7

30.6

56

6 10

13

37

5 18

.5

24.0

9.

9 20

.4

9 29

.3

46.9

21

2 1

8 74

4.

5 3.

7 0.

7 2.

6

10(m

) 25

.3

20.0

3

0 2

3 0.

2 0.

3 0.

6 0.

3

11

32.3

47

.4

390

4 8

156

8.2

8.2

0.4

5.5

12

25.5

47

.0

246

1 11

84

5.

2 2.

6 3.

7 3.

0

13

28.7

8.

5 1

1 6

6 0.

1 -

- 1.

2

14

26.2

8.

5 13

9 0

4 53

16

.3

- -

10.4

15

28.2

47

.1

577

24

25

588

12.3

25

.4

11.1

20

.8

16(m

) 36

.1

21.7

16

1 7

12

125

7.4

12.5

10

.1

9.6

Med

ian

29

.0

22.0

22

0.0

2.5

10.5

11

0.0

10.2

8.

4 6.

3 10

.5

39

Of the four symptomatic patients, two patients could not discontinue their PPI-treatment due to the severity of their symptoms. The percentage of time with pH<4 was found to be 0% in both these patients. The remaining two patients interrupted their PPI-medication two weeks prior to the Bravo-capsule placement. Their plots showed a percentage of time with pH<4 of 11.3 and 10.0%. A small stomal ulcer was seen at endoscopy in one of these patients.

Symtoms PPI BMI Registration

Time (h) Episodes

pH<4 pH<4

>5 min Time

pH<4(min) Episodes

pH<4/hour % time pH<4

Epigastralgia Discont 2

wks 39.3 21.2 63 8 143 3.0 11.3

Epigastralgia/ stomal ulcer

Discont 2 wks

33.0 45.5 421 6 275 9.3 10.0

Epigastralgia On demand 24.7 44.7 5 0 1 0 0.0

Epigastrlgia Continously 33.1 38.8 0 0 0 0 0

Table 3 pH in the gastrojejunostomy in symptomatic patients after RYGBP

No complications occurred during placement or passing of the capsule. One patient reported mild retrosternal discomfort when returning the receiving device (two days after placement), but an additional endoscopy was normal and the discomfort soon subsided.

40

Paper III

Gastric emptying

All subjects could ingest the meal within the stipulated two minutes. Data were extracted at 5, 10, 15, 20, 30, 45, 60, 75, 90 and 120 minutes and the mean percentages of remaining tracer in the gastric tube were plotted. The mean half-emptying time was 28±16 minutes, and areas of normal lag-phase and half emptying time used in our lab133 have been included in the figure to give an idea of gastric emptying in normal subjects (Figure 10). A lag-phase was seen in six patients (30%). In the remaining patients, the emptying was linear already from the first scan (5 minutes).

Figure 10 The mean percentage of remaining tracer in the gastric tube after ingestion of the 99mTc-labeled omelette (77.5 kcal) for all study patients is plotted. Bars show the standard error. For comparison, the areas of the normal lag-phase and half-emptying time in our lab after a standard meal (310 kcal) in normal subjects are indicated with ovals

PYY

The fasting PYY-levels were significantly higher in the DS-group compared to the non-operated age-matched controls (p<0.001). Thirty minutes after the

41

meal, PYY had risen from a mean of 161 to 277 pg/ml, an elevation of 72%. PYY-levels were continuously high in the DS group during the study period, with a declining trend three hours after the test meal. In the controls, a slow and continuous rise was seen postprandially (Figure 11).

Figure 11 PYY levels in ten DS patients (squares, solid line) and nine age-matched controls (diamonds, dotted line). Time is given in relation to the test meal, consumed after an overnight fast. Note the marked response to the test meal as well the con-stantly elevated PYY levels in the DS-patients. p<0.001, ANOVA repeated meas-ures analysis. Bars show the standard error

Dumping No patient reported daily dumping symptoms. One patient reported weekly dumping symptoms, two patients monthly and eight patients once or a few times a year. The remaining nine patients reported that they never experi-enced dumping. We found no correlation between reported dumping symp-toms and gastric emptying times.

42

Paper IV

Perioperative All patients underwent surgery as planned (randomized or chosen proce-dure). The outcome was similar between randomized and non-randomized patients, in both types of operation. In short, DS-patients had longer duration of surgery (156 ±30 vs. 117 ±33 min, p<0.001), higher mean postoperative C-reactive protein levels (138.5 ±54.7 vs. 117.6 ±43.2, p=0.047) and higher total morphine consumption postoperative day one to three (133.3 ±84.8 mg vs. 89.3 ±72.9 mg, p=0.010) than RYGBP-patients (n=90). The resume of oral intake; liquids (2.2 ±2.8 vs. 1.2 ±0.5 days, p<0.001, n=90) and solids (4.4 ±1.1 vs. 3.3 ±0.8 days, p<0.001, n=90) was delayed after DS compared to RYGBP. The average length of stay was 7.0 ±4.1 and 5.2 ±1.2 days, p=0.004, n=90, for DS and RYGBP, respectively (Table 4).

In total, three reoperations were made (two DS, one RYGBP). One DS patient was drained due to a bile collection without obvious source and she could be discharged ten days postoperatively. The other DS patient was reo-perated due to postoperative bleeding causing elevated intraabdominal pres-sure and leakage at the divided duodenal bulb. After drainage, this subsided and the patient could be discharged after 32 days. One RYGBP patient had a negative laparotomy on the first postoperative day due to a suspected leak and had an uneventful recovery. One patient died, due to pulmonary embol-ism, occurring in her home three weeks after a DS, in spite of a previous uneventful postoperative course.

Follow up A follow-up of 90% was achieved and the median time from operation was 3.5 (2.0-5.3 years). The weight loss was significantly greater in the DS-group (24.0±6.8 vs. 17.3±6.1 BMI-units, respectively, p<0.001). In the RYGBP group, 10 percent of the patients had a weight reduction of less than 10 BMI-units. This was not seen in the DS-group (Figur 12). The percentage excess weight loss (EWL%) was 81.1% in the DS group, and 63.2% in the RYGBP group (p<0.001). 91.1 % of the DS patients and 75.6% of the RYGBP-patients achieved an EWL of more than 50% (p=0.004). The BMI-reduction was independent of whether the patients operation had been ran-domized or not.

Tab

le 4

Per

iope

rativ

e da

ta, †

Stu

dent

s t-

test

, ‡ M

ann

Whi

tney

U te

st, ¶

Fis

her

exac

t tes

t

R

and

omiz

ed p

atie

nts

n

=47

P

roce

du

re o

f p

atie

nts

ch

oice

n

=43

T

otal

n

=90

D

S, n

=24

R

YG

BP

, n=

23

D

S, n

=21

R

YG

BP

, n=

22

D

S n

=45

R

YG

BP

, n=

45

Dur

atio

n of

sur

gery

(m

in)

153.

2±23

.6

117.

4±40

.7

p<0.

001†

16

0.2±

36.6

11

6.2±

22.3

p<

0.00

1†

156.

5±30

11

6.8±

32.7

p<

0.00

1†

Mea

n C

RP

day

1-3

(mg/

L)

137.

8±59

.3

125.

0±50

.8

p= 0

.433

† 13

9.4±

50.5

11

0.0±

32.4

p=

0.02

8†

138.

5±54

.7

117.

6±43

.2

p=0.

047†

Mor

phin

e-co

nsum

ptio

n da

y 1-

3 (m

g)

140.

4±96

.3

92.9

±60.

7 p=

0.05

0 ‡

125.

2±71

.0

85.5

±85.

1 p=

0.10

5‡

133.

3±84

.8

89.3

±72.

9 p=

0.01

0‡

Tim

e to

liqu

ids

per

mou

th (

days

) 2.

8±3.

8 1.

3±0.

6 p<

0.00

1 ‡

1.6±

0.7

1.1±

0.3

p<0.

015‡

2.

2±2.

8 1.

2±0.

5 p<

0.00

1‡

Tim

e to

foo

d pe

r m

outh

(day

s)

4.5±

1.3

3.5±

0.9

p<0.

003‡

4.

4±0.

7 3.

1±1.

0 p<

0.00

1‡

4.4±

1.1

3.3±

0.8

p<0.

001‡

Len

gth

of s

tay

(day

s)

7.6±

5.4

5.5±

1.2

p=0.

073 ‡

6.

4±1.

6 4.

9±1.

1 p<

0.00

1‡

7.0±

4.1

5.2±

1.2

P<

0.00

4‡

Mor

tali

ty

1 0

p=0.

511¶

0

0 -

1 0

p=0.

500¶

Reo

pera

tion

s 2

1 p=

0.51

0 0

- 2

1 p=

0.50

Pos

tope

rati

ve tr

ansf

usio

ns

0 2

p=0.

234¶

2

1 p=

0.48

2 3

p=0.

500¶

44

Figur 12 Histogram of BMI-reduction after DS (blue columns) and RYGBP (red columns. p<0.001 Students t-test

In the DS group, ten of the eleven patients with diabetes had been able to stop oral anti-diabetic medication and two of thirteen had discontinued their antihypertensive treatment. Of the three diabetics in the RYGBP-group, none had medication at follow up and four of fifteen had discontinued the antihy-pertensive treatment.

No base-line differences between the two operations were seen in the la-boratory tests. DS-patients had significantly lower glucose (4.64 ±0.44 vs. 5.29 ±1.04 mmol/L, p=0.008) and HbA1c (3.97 ±0.34 % vs. 4.67 ±0.32 %, p=0.005) tree years postoperatively compared to RYGBP (Figure 13).

Figure 13 DS yielded significantly lower HbA1c levels, p=0.005, ANOVA repeated measures analysis. Bars denote 95 % confidence interval

45

At one year, one DS-patient (2.4%) and four RYGBP-patients (9.5%) had a fasting glucose above 6 mmol/l (p=0.187). The corresponding proportions at three years were 0 and 12% respectively (p=0.083). No differences in hae-moglobin or albumin were seen. Nor was the frequency of anaemic or hy-poalbuminemic patients different between DS and RYGBP (data not shown). Improvements were seen in folate and vitamin B12-values in both groups, probably due to supplements and diet counselling (Table 5).

Concerning abdominal symptoms, diarrhea and malodorous flatus were more frequent after DS, and mostly occurred weekly, while dumping was more common after RYGBP (Figure 14). On the effect of the operation on their general well being, 100% of the DS-patients, and 85% of the RYGBP-patients responded that they were satisfied or very satisfied in an ordinal scale of four steps (appendix 1) (p=0.086). The five RYGBP-patients who were dissatisfied (n=5) or very dissatisfied (n=1) had lower weight loss, 8.4 BMI-units (1.8-13.0) as compared to the whole RYGBP group (p=0.002).

Figure 14 The DS group is represented by X and the RYGBP group by O. Abdo-minal symptoms were evaluated on an ordinal scale with five steps. N=never. Y=yearly symptoms, M=monthly symptoms, W=weakly symptoms and D=daily symptoms (indicated on the bar on the bottom right). These answers were recoded to numbers (0-4) and means calculated for visual representation only. p-values are calculated non-parametrically (Mann-Whitney-U-test)

46

Table 5 Laboratory follow up after DS and RYGBP. ANOVA repeated measures analysis is used

DS RYGBP

Hemoglobin (g/L) Preoperatively One year postop Three years postop

141±13 (n=45) 135±14 (n=40)

128.7±14 (n=39)

139±11(n=44) 138±15 (n=42) 133±16 (n=35)

p=0.280

Kobalamines (pmol/L) Preoperatively One year postop Three years postop

283±130 (n=45) 491±370 (n=40) 478±321 (n=35)

280±107 (n=43) 403±297 (n=44) 478±280 (n=30)

p=0.458

Folate (nmol/L) Preoperatively One year postop Three years postop

9.5±5.9 (n=45)

21.3±15.7 (n=40) 67.3±205 (n=38)

11.0±9.2 (n=43)

16.8±11.9 (n=44) 61.0±176 (n=28)

p=0.977

Albumine (g/L) Preoperatively One year postop Three years postop

38.2±2.8 (n=45) 38.3±3.1 (n=42) 38.5±3.3 (n=35)

39.0±2.4 (n=44) 37.8±2.8 (n=45) 38.6±3.3 (n=22)

p=0.363

Glucose (mmol/L) Preoperatively One year postop Three years postop

6.43±3.35 (n=43) 4.56±0.45 (n=41) 4.64±0.44 (n=35)

5.74±1.20 (n=44) 4.94±0.68 (n=42) 5.29±1.04 (n=25)

p=0.008

HbA1c (%) Preoperatively One year postop Three years postop

5.24±1.27 (n=38) 3.90±0.42 (n=42) 3.97±0.34 (n=30)

4.89±0.87 (n=37) 4.41±0.39 (n=43) 4.67±0.32 (n=17)

p=0.005

HDL (mmol/L) Preoperatively One year postop Three years postop

1.32±1.21 (n=44) 1.07±0.24 (n=42) 1.17±0.37 (n=32)

1.14±0.23 (n=43) 1.29±0.37 (n=44) 1.34±0.35 (n=17)

p=0.149

LDL (mmol/L) Preoperatively One year postop Three years postop

2.83±0.89 (n=44) 1.90±0.64 (n=42) 2.00±0.69 (n=28)

2.93±0.69 (n=44) 2.36±0.52 (n=44) 2.60±0.63 (n=14)

p=0.057

Triglycerides (mmol/L) Preoperatively One year postop Three years postop

2.38±1.84 (n=44) 1.18±0.75 ((n=42) 1.00±0.51 (n=32)

1.82±0.76 (n=44) 1.11±0.35 (n=44) 1.15±0.62 (n=17)

p=0.138

47

During the follow up, one DS-patient had undergone coeliotomy due to in-ternal herniation with incarcerated small intestine and another had undergone cholecystectomy. In addition, nine DS patients had been readmitted due to abdominal pain (whereof five were diagnosed with cholecystolithiasis). In the RYGBP-group, one patient had been operated due to small intestinal obstruction and another had undergone cholecystectomy. Five additional readmissions had been made due to unclear abdominal pain. Five patients (DS, n=3 and RYGBP, n=2) had undergone incisional hernia repair during the follow-up period.

48

General discussion

Continuous evaluation of all surgical methods is vital to optimizing patient treatment. Bariatric surgery is no exception. In this thesis we examine as-pects of the postoperative gastrointestinal function in relation to complica-tions, symptoms and metabolical markers after weight reducing surgery.

In Paper I, the role of hydrochloric acid in patients with stomal ulcer is evaluated. Stomal ulcers are known to occur after gastric resections, and in frequencies of 0.6-16% after RYGBP95, 97, 99, 100, 102-104, 134. The pathogenesis of stomal ulcer has been under some debate. Gastric acid is produced mainly in the fundus and corpus of the stomach where parietal cells are abundant. Cardiac mucosa has been considered to be solely mucus producing and thus devoid of acid output135. The proximal pouch in RYGBP is constructed high on the lesser curvature, and measurements with a perfused orogastric tube has failed to detect any significant amounts of acid in an intact proximal pouch111. Some authors have therefore ascribed gastrogastric fistulation, when acid from the bypassed stomach gains entrance to the proximal pouch, to be the mechanism whereby the Roux-limb can be exposed to acid104. However, autopsy studies have challenged the existence of gastric muco-sa136, and reports from our group demonstrate presence of parietal cells all the way up to the gastroesophageal junction112.

The main finding in paper I was the more acidic environment in stomal ulcer patients as compared to asymptomatic controls. Indeed, the ideal anas-tomosis for avoiding stomal ulceration would be an esophagojejunostomy, but this gives, in my opinion, an unacceptably high risk of anastomotic lea-kage, which is reported in up to 11 % after gastrectomy for malignancy137. Therefore, all efforts to minimize the pouch size should be made without risking the reliability of the anastomosis. After the pH-measurements, three of the patients with stomal ulcer were reoperated with reduction of their proximal pouches in order to reduce the amount of acid producing parietal cells. In two of these, discontinuation of PPI and resolution of symptoms were obtained, but one patient relapsed with a bleeding stomal ulcer and life-long PPI was prescribed. The anastomosis and the ulcer itself were not re-sected in the patient who relapsed, and one could speculate that this contri-buted to the relapse.

All patients in this study had discontinued their PPI for at least one week prior to the investigation. A weakness of this design can be an acid rebound seen in response to PPI-withdrawal138. The majority of the patients, however,

49

had only had PPI for a short time and I do not believe that this explains the whole difference in acidity demonstrated in this study.

Interestingly, some patients seem to be impervious to acid exposure in the Roux-limb since not all patients with gastro-gastric fistulation develop stom-al ulceration even though they have a low pH in their proximal pouch104. This notion is also supported by the fact that early routine endoscopy reveals a high frequency of stomal ulceration (7.9%) one month after surgery and only 0.3% one to two years postoperatively102. There are no data on how many of the “early” stomal ulcer patients who were under PPI-treatment at the later follow up point in this study, but it seems likely that the jejunal mucosa is more susceptible to acid in the early postoperative period, and that an adaptation occurs over time. Lublin et al. advocate a routine 30-day PPI-course in order to avoid these early ulcerations, but still, late ulcerations appear and they seem prone to perforate101. It is worth notice that all eight perforated stomal ulcers in this series arose on antecolic Roux-limbs in a patient material where both antecolic (n=499) and retrocolic (n=403) Roux-limbs were included (p<0.05). Furthermore, perforated stomal ulcers seem to have an intriguingly consistent location at the top of the antimesenteric side of the Roux-limb101, 139, which could imply a circulatory facet to the ulcero-genesis.

In paper II, we adopted the wire-less BRAVO-system™ to study the pH in the proximal jejunum after RYGBP. We were able to present long-time reg-istrations (exceeding 24h) of pH in the proximal Roux-limb for the first time. Certain modifications in the placement of the capsule gave longer registra-tion times. I chose to place all capsules under direct visual control with the endoscope to be sure of their position. In some patients, the egress from the pouch to the Roux-limb was not straightforward, and in these cases the en-doscope was very useful to guide the applicator in the right direction, in or-der to position the pH-sensor in the jejunal lumen. The positioning-hole of the BRAVO™-capsule was ideally placed on gastric mucosa for a more reliable placement. During the study, I also increased the time with applied suction to two minutes instead of one. With these measures, a registration of more than 24 hours was generally obtained (mean 25.7 hours).

We found a median of 10.5% of time with pH<4 at the level of the gastro-jejunostomy in the 16 asymptomatic patients with successful recordings. We adopted the cut-off level at pH<4, used in the evaluation of gastroesophageal reflux disease, for several reasons. First, a pH-level below 4 gives rise to heartburn in normal subjects140 suggesting an erosive effect on mucosa nor-mally only subjected to neutral pH. Second, normal subjects without heart-burn have a pH greater than 4 in 98,5% of the time in a 24 hour registration141. Third, and maybe most relevant to the present study, pepsino-gen is activated into pepsin, which plays a fundamental role in formation of ulcers in the gastrointestinal tract, below a pH of 4142. Fourth, in an experi-

50

mental setting, jejunal ulcers have been produced with pepsin and hydroch-loric acid 142, 143. This is, in our opinion the most plausible pathogenesis for stomal ulcers seen after RYGBP and further studies of pepsin activation in the gastrojejunostomy are warranted.

The fact that asymptomatic patients can have as much as 37.7% of time with pH<4 without ulcer formation or symptoms is interesting and in line with the before mentioned report of patients without ulcers but low pH be-cause of gastrogastric fistulation 104. Our data from paper II is also in line with the acidity found in the proximal pouches of the control group in paper I (6.7-43.9 % of time with pH<4). The conclusion that pH is closely related to ulcer formation is supported by the results in paper I, but the nature of the defense mechanisms, which seem to develop in the jejunal mucosa of the Roux-limb, remains to be completely understood. Smoking and non-steroid anti-inflammatory drugs have been linked to stomal ulceration144. Helicobac-ter pylori infection, even if successfully treated could also predispose for stomal ulcer formation97 but other authors have not found this association145. Other microbial factors of this normally sterile locale might play a role here, as might histological changes that occur after RYGBP146, but this has not been addressed from an acid-protection point of view and further study of this would be interesting.

We also investigated four symptomatic patients. Of these, two had dis-continued their treatment for two weeks, and their pH-results were in line with the asymptomatic group (10.0 and 11.3% of time with pH<4), thus no obvious rebound phenomenon could be seen. Two patients were studied during PPI-therapy and in these, no time with pH<4 was registered at all, which is reassuring considering the wide-spread use of PPI.

In paper III rapid gastric emptying after DS, with a mean half-emptying time of 28 minutes, is demonstrated. Gastric emptying after DS is not previously described. In this study, we aimed primarily at describing the postoperative gastric emptying since normal gastric emptying in normal weight and obese subjects are described prieviously132, 147, 148. In order to achieve this, we re-duced the solid meal to about 80 ml (75 kcal) since studies of the stomach remnant after DS have demonstrated volumes as low as 80ml149. With the hypothesis of fast emptying, we also stipulated that the subjects should in-gest the meal in two minutes. The half emptying time of the gastric tube with this method was 28±16 minutes. There are two studies in the literature where emptying of the gastric tube after sleeve gastrectomy is investigated150, 151. Different protocols are used for scintigraphic studies in the literature, why comparisons of our results must be made with great caution. Braghetto22 et al used an egg sandwich with approximately 200 g of bread, consumed within ten minutes, and obtained a postoperative half-emptying time of 38 minutes, while Bernstine 23 et al found a postoperative half emptying time of 57 min-utes using a 230 kcal meal consisting of bread (60 g) and scrambled eggs

51

(35% fat, 47 % carbohydrates and 18% protein, ingestion time not speci-fied). We interpret our result of 28±16 minutes as being quicker than after sole sleeve gastrectomy. This could be explained by denervation during the dissection around the pylorus, which takes place in the duodenal switch part of DS.

The normal lag-phase113 was absent in all but six of the study subjects. The definition of the lag-phase in gastric emptying has been under some debate152. In this setting of postoperatively deranged anatomy and pylorus function, a description of emptying cannot readily be compared to normal gastric emptying. It seems illogical to attribute any of the emptying curve as lag-phase when it is essentially linear from the first (five minutes) scan, even if the crossing of the regression line at 90% is normally used for lag-phase definition in our lab132. We have therefore studied the graphs individually, and designated a lag-phase in those patients who demonstrated any biphasic emptying pattern. Indeed, the frequent lack of lag-phase and quick gastric emptying is also described after truncal vagotomy and pyloroplasty for ulcer disease153. There was nor correlation between gastric emptying rate and weight reduction, nor to time since surgery (data not shown).

In the fasting state, higher levels of circulating PYY were found in the DS-group compared to the controls. Elevated PYY levels have also been reported in jejuno-ileal bypass patients129, having a similar alternation of the small bowel. In contrast, non-operated obese subjects have low124 or normal125-127 levels of PYY. In the present study, a marked PYY response to the test meal was also seen in the DS-group. This finding, as well as our earlier report of an exaggerated PYY-response in gastric bypass patients compared to controls131, are likely due to the rapid appearance of nutrients in the ileum. Active PYY3-36 has high affinity to Y2-receptors in the hypo-thalamus, inducing satiety in humans124. Even though its effect on food in-take in humans later has been disputed128, one might speculate that in addi-tion to the restrictive and malabsorptive effects of DS, the patients also bene-fit from a hormonal mechanism with constantly elevated PYY and rapidly rising levels at meals.

Disturbing dumping symptoms were surprisingly uncommon in our DS-patients. In other procedures with swift passage to the small bowel (to-tal/subtotal gastrectomies and gastric bypass) dumping can be a considerable and long-lasting problem. In a small study of six patients with dumping problems after gastric surgery, increased PYY-levels in response to orally administered glucose was reported154. In our study, however, dumping was uncommon and the now demonstrated reservoir function of the gastric tube might explain this. In accordance, postprandial proinsulin, insulin and glu-cose levels in BPD-DS patients are similar to those of normal-weight control subjects, as earlier demonstrated155. This is not the case after gastric bypass, where ingested food immediately reaches the small bowel, giving rise to a high glucose and insulin levels, as well as an elevated frequency of dumping.

52

However, the relatively long time after surgery (3.5 years) might also have given our BPD-DS patients a chance to adapt their eating behaviour to avoid dumping.

In paper IV, the aim to achieve a prospective controlled randomized trial, was only partly met. Since both operations were part of the treatment arsen-al, we complied with patients who wanted DS outside the study. As time passed, it became obvious that as much as half of the patients chose their type of operation themselves. These decisions were often influenced by suc-cessful cases in their surroundings and since we aimed at randomizing as many as possible, we did not favor either procedure. Since the patients choosing DS turned out to be well matched with those choosing RYGBP, we decided to do a follow up of these patients as well. In this way we feel that as much information as possible is extracted from our experience.

Our BMI-cut-off at 48 units was chosen in order to include 80 patients within three years based on our referral rate. Some might argue that this is a low BMI for DS and indeed, some authors recommend lap-RYGBP as the standard operation in this group of patients156. On the other hand, Hess et al, with extensive experience of over 1300 cases, prefer the DS in all morbidly obese patients regardless of BMI75. Furthermore, other criteria are also im-portant (distribution of fat, diabetes, gender) in selecting patients for a spe-cific type of operation.

In this study we chose to use our standard technique in super obese pa-tients, the open approach. First, when enrolment began, no standard proce-dure for laparoscopic DS existed and second, we did not want a steep learn-ing curve to interfere with the results. The overall complication level in the study was indeed low and we believe that three reoperations (bile leakage, haemorrhage, negative exploration) in a patient cohort like this is acceptable. Kim et al has compared open and laparoscopic DS, with a mortality of 7.6% in the laparoscopic group106 and in the literature, mortality after DS varies from 0-3.5%75-77, 106, 157. Our only death, due to pulmonary embolism, oc-curred in the DS-group (2.2%). Fatal pulmonary embolism is a very rare occurrence and hard to ascribe to the specific type of operation. Duration of surgery does however, play a role in the risk for tromboembolic complica-tions.

The duration of surgery was longer in the DS group. This is not surprising since DS implies a larger operative field, with extended dissection and a hand-sutured anastomosis. However, our mean operating time, 153 minutes for DS, is shorter than earlier reported (259 minutes)106. The length of stay as well as the time to ingestion of fluids and solids were longer after DS. The more extensive operation will most likely lengthen the postoperative ga-strointestinal paralysis, thus prolonging the time requiring nil per mouth. Our demand that all patients must to be able to eat normal food before discharge inevitably results in an extended length of stay after DS and a longer length

53

of stay after DS is in accordance with the literature79, 158, 159. The significantly higher levels of C-reactive protein and morphine consumption during the first three postoperative days after DS also reflect the more complex nature of this operation.

The major finding was the large difference in BMI-reduction, 24 BMI-units and 17 BMI-units, for DS and RYGBP, respectively. This is in line with earlier studies comparing the two procedures76-78. DS can therefore be recommended to super-obese patients who need massive and reliable weight loss. A sizable portion of the super obese patients obtains dissatisfactory weight results from RYGBP. Eleven of the RYGBP-patients in our study (24%) had not achieved a 50% excess weight loss, which is a common measure for successful bariatric surgery, as compared to 4 patients (8.9%) in the DS group. Moreover, the RYGBP group included five patients (12.5%) who had a weight loss of less than 10 BMI-units, whereas the lowest weight loss in the DS group was 12.1 BMI-units. Further treatment of the patients who fail to loose weight after bariatric surgery or regain weight over time represents a big and increasing challenge to the medical community.

Lower glucose and HbA1c levels were found after DS and after three years, no hypergycemia was found in the DS-group. Recently, DS-patients were demonstrated to have fasting glucose levels below that of normal weight control subjects after DS155. Apart from superior glucose control, a high resolution-rate of other comorbidities (diabetes, hypertension and hypercholoesterolemia) are reported after DS20.

No symptomatic hypoalbuminemia occurred in either group. This is oth-erwise a rather common complication after DS, especially in patients with a short common channel. We can confirm earlier reports stating that hypoal-buminemia is rare using a constant 100-centimetre common channel in all patients20. It is, nevertheless, our clinical impression that an elevated suspi-cion of nutritional defects is warranted, and since normal multivitamin subs-titution may be inadequate even in RYGBP160 and a greater risk of vitamin A- and D-deficiencies after DS has been reported90, additional supplements could be advantageous in this more malabsorptive procedure.

The risk of gastrointestinal problems is important for the patient. Diarrhea and malodorous flatus is a more common problem after DS, even though these symptoms can occur after RYGBP as well. Higher frequency of bowel movements after bariatric surgery are reported prieviously161, 162 and some improvement over time is reported157. Wasserberg et al, however, found no difference in bowel function between DS and RYGBP163. Soiling, a social handicap, was rare in our material. As expected, RYGBP was more prone to give rise to dumping symptoms. Neither heartburn nor vomiting was a com-mon problem. This is to be expected since none of the operations have a pronounced restrictive component. Overall, patients reported a very high level of satisfaction, 100% of the DS-patients, and 85% of the RYGBP pa-tients, were satisfied or very satisfied (p=0.086). As mentioned before, non-

54

satisfied patients had lost less weight than the RYGBP-group as a whole, again implying that the weight result is of utter importance for quality of life.

One drawback of the present study is the rising number of patients that declined randomization between the two operations. There were no differ-ences in weight or age between the two groups and we believe that the weight result and metabolic parameters are unlikely to be affected by the patients' choice. The subjective symptom-scores could be influenced by an-ticipation, but the overall results of the randomized and non-randomized patients were similar. Randomization aside, the operations were not blinded for the patient, and the experience of gastrointestinal symptoms could be influenced by expectation in relation to the performed procedure as well.

In summary, DS produces superior weight loss and better glucose control compared to RYGBP in the long run, with acceptable nutritional risks and without intolerable complications for the patient. We believe that DS has a place in future surgical treatment of super obese patients, an increasing pa-tient cohort, as well as in surgical treatment of failed RYGBP.

55

Conclusions

Paper I Patients with stomal ulcer after RYGBP, have lower pH in their proximal pouch as compared to asymptomatic control subjects. Acid seems to play a fundamental role in stomal ulcer formation. Reduction of the pouch size did not eliminate stomal ulcers completely.

Paper II Wire-less pH monitoring in the proximal Roux-limb after RYGBP is feasible and safe. The mucosa of the proximal Roux-limb is exposed to pH below four in median of 10.5% of the time even in asymptomatic patients.

Paper III Although the pylorus is preserved in DS, gastric emptying is fast. After DS, PYY levels are elevated in the fasting state as well as after a standardized meal compared to controls. The marked postprandial response could be caused by the rapid appearance of nutrients in the distal ileum and despite this, dumping symptoms are uncommon.

Paper IV DS produces superior weight results and glucose control as compared to RYGBP in patients with BMI>48. Both operations are safe and yield high satisfaction-rates but a higher risk of diarrhea and malodorous flatus arise after DS.

56

Svensk sammanfattning (summary in Swedish)

Övervikt och fetma är växande problem i Sverige och globalt. Detta innebär att en allt större del av befolkningen riskerar sänkt livskvalitet, sjukdom eller för tidig död till följd av detta. Ett stort antal behandlingsmetoder för be-handling av fetma har provats, men den enda behandling med varaktigt re-sultat på vikt och följdsjukdomar är kirurgi. Gastric bypass (RYGBP), som innebär att magsäcken ”kopplas ur” och att en liten magsäcksficka kopplas direkt till tunntarmen, är den i särklass vanligaste operationsmetoden i Sve-rige idag. Av de överviktiga är det gruppen med de allra tyngsta som ökar mest och i denna grupp har gastric bypass sämre resultat. Vi har sedan 2003 därför även använt biliopankreatisk divergering med duodenal switch (DS), för att hjälpa patienter med BMI>48. Både RYGBP och DS har specifika för- och nackdelar och denna avhandling studerar de förändringar i mag-tarmkanalen som uppkommer efter överviktskirurgi.

Delarbete I I detta arbete studeras en av de vanligaste komplikationerna efter gastric bypass; stomalt sår som uppkommer i den tunntarmsslynga som kopplas till den lilla magsäcksfickan. Varför vissa patienter får denna komplikation vet man ej. Man har tidigare inte kunnat visa någon signifikant syraproduktion i den lilla magsäcksfickan och därmed ansett att gastro-gastrisk fistel (då salt-syra från den urkopplade magsäcken kommer över till fickan), ligger bakom all sårbildning. Huvudfyndet i delarbete I är ett mycket lågt pH i den lilla magsäcksfickan vid mätning med nasogastrisk pH-givare hos patienter med stomalt sår, jämfört med symtomfria kontroller. Detta låga pH förelåg trots att gastro-gastrisk fistel uteslutits med röntgen. Tre av patienterna i studien opererades med förminskning av magsäcksfickan och av dessa blev två sym-tomfria. Den tredje fick dessvärre ett nytt sår och får fortsätta med livslång medicinsk behandling.

Delarbete II Tekniken för trådlös pH-mätning i magtarmkanalen, det så kallade BRA-VO™-systemet, anpassas i detta arbete till att undersöka surhetsgraden i det område där stomala sår bildas. Gastric bypass-opererade patienter utan sym-tom på stomalt sår undersöktes genom att man med hjälp av gastroskop pla-

57

cerade en BRAVO™-kapsel vid tunntarmens koppling till magsäcksfickan och på detta sätt kunde pH registreras i upp till 48 timmar medan patienten levde som vanligt. Metoden fungerade och kapslarna kunde placeras i önskat läge under ögats kontroll med hjälp av gastroskopet. Tekniken för att fästa kapseln modifierades och allt längre registreringstider erhölls i och med det-ta. Med denna metod fann vi en relativt sur miljö, med ett pH<4 under 10.5% av registreringstiden. Fyra undersökningar av patieneter med symtom gjordes också. Två undersöktes under pågående syradämpande behandling och de hade aldrig lägre pH än 4. Resultaten visar en signifikant produktion av magsyra trots att magsäcksfickan görs liten. Att de flesta patienter trots detta inte utvecklar sår talar för att en adaptation av slemhinnan i tunntarmen sker.

Delarbete III Vid DS görs magsäcken om till ett smalt rör. Nedre magmunnen bevaras vid denna operation, men dess funktion efter operationen är inte undersökt. Ett stort antal hormoner är inblandade i reglering av magsäckstömning och många av dessa har också hungerreglerande funktioner. I detta arbete foku-serar vi på PYY, som hos icke-opererade personer har betydelse för mag-säckstömning och hungerupplevelse. Vi undersökte magsäckstömning på 20 patienter efter DS med hjälp av en radioaktivt märkt omelett. På tio av dessa undersöktes också PYY-nivåer i fasta och efter en standardiserad måltid. Vi utvärderade också det symtomkomplex som kallas dumping och som ofta förekommer efter magsäcksoperationer. De fynd som gjordes kan samman-fattas med en snabb magsäckstömning, sannolikt till följd av att nedre mag-munnens nervförsörjning påverkats av operationen, höga PYY-nivåer och få dumpingsymtom. Även om magsäckstömningen var snabbare än normalt, så var den inte momentan utan hade en halveringstid på 28 minuter i genom-snitt. Detta förklarar att dumping är sällsynt efter denna operation. De höga PYY-nivåerna är ett intressant fynd och man kan spekulera i om dessa nivåer har en särskild effekt på mättnadskänslan. Det skulle kunna bidra till den goda vikteffekt som uppnås med DS på lång sikt.

Delarbete IV Superobesitas (BMI>50) är den snabbast växande patientkategorin bland de sjukligt överviktiga i USA. Gastric bypass ger inte optimalt viktresultat hos dessa patienter. Vi har i en randomiserad studie mellan DS och RYGBP jämfört perioperativa risker, viktresultat och gastrointestinala symtom. Ut-värderingen är gjord i median 3,5 år efter operationen och innehåller även biokemiska jämförelser. Huvudfyndet är en kraftigare viktnedgång efter DS (24 BMI-enh) jämfört med gastric bypass (17 BMI-enh). Duodenal switch ger mer besvär av diarréer och illaluktande gaser, medan gastric bypass ger

58

dumping. Större förbättring av glukos- och HbA1c-nivåer förelåg efter DS. Patienterna är mycket nöjda med effekten av operationen på deras allmänna välbefinnande efter både DS och RYGBP. De viktresultat vi kan visa är i nivå med de flesta andra studier. Att DS har en mycket god effekt på diabe-tes är känt, men skillnaden i fasteglukosvärden har inte hittats tidigare.

Konklusioner I Trots att magsäcksfickan görs liten finns en signifikant syraproduktion

kvar vilket kan leda till mycket låga pH och stomala sår i tunntarmens första del.

II Den trådlösa BRAVO-kapseln kan användas för att undersöka patienter och mäta pH kontinuerligt under två dygn efter gastric bypass. Den för-sta delen av tunntarmen utsätts för låga pH-värden även hos symtomfria patienter, vilket antyder att en adaptation av slemhinnan sker över tid ef-ter operationen.

III Tömningen av magsäckstuben efter duodenal switch är snabb, men inte momentan vilket kan förklara att så få patienter får dumpingsymtom postoperativt. PYY-nivåerna är höga i fasta och stiger kraftigt efter mål-tid jämfört med åldersmatchade kontroller.

IV Duodenal switch ger större viktnedgång och lägre blodsockernivåer än gastric bypass vid BMI>48. Detta sker till priset av mer besvärande diar-réer och illaluktande gaser. Riskerna vid de två ingreppen är låga och likvärdiga.

59

Acknowledgements

To me, the greatest merit of this work is all the bright and inspiring people it has helped me to befriend. I could have done none of this myself and I am grateful to many. My special thanks however, go to:

Magnus Sundbom, my main tutor and good friend, for tireless revisions of my papers and unending drive in scientific and clinical work. I am also thankful for your willingness to share your deep knowledge in the field of bariatric surgery, cancer surgery and science, and for having confidence in me. Thank you for helping me in many tricky surgical situations. I look for-ward to performing many challenging operations together in the future.

Sven Gustavsson, former head of bariatric surgery and my first scientific tutor, for inviting me into his team and generously sharing his vast know-ledge.

Ulf Haglund, my co-tutor and professor in surgery for providing the possi-bility for this work and valuable advice in research matters as well as in dif-ficult clinical situations.

Ewa Lundgren, Staffan Wollert and Claes Juhlin, former and present heads of the surgical department and Lars Wiklund, head of the department of surgical sciences for providing the means and facilities for academic sur-gical activities.

All my other co-authors: Hans Hedenström, Anders Karlsson, Britt-Edén-Engström and Sven Nilsson for valuable contributions and interesting discussions.

All the other former and present members of the bariatric surgery team, Anders Ahlström, Zakaria Abdulla, David Edholm, Annika Norstedt, Annika Dannberg, Anita Ohlin, Hanna Hellsten and many more for pro-viding a positive and inspiring work-place and caring for our patients. Anna Ehrenborg for all this, but also for her dedicated work on our bariatric data-bases and invaluable help in recruiting patients.

60

All other colleagues at the department of surgery, and especially the upper GI-team, for providing support and a stimulating and creative working envi-ronment.

All colleagues and co-workers in Gävle among whom I would like to give special thanks to Michael Krogh, Jörn Holm and Torbjörn Tuvesson for being fantastic teachers and role-models early in my surgical career. Mats Savlid for a great introduction to surgery whilst commuting, and Dan-Olle Westergren for great companionship in the tough early years of our careers.

Erik Mörtberg for great friendship and technical support.

All coworkers at the department of endocrinology, led by professor Anders Karlsson, for all inspiring seminars over the years and Margareta Eriksson who in addition to skillful labwork cook expertly.

The staff at our endoscopy outpatient clinic led by Ulrika Dovner, for unbe-lievable flexibility and positive attitude when I want to to “just one more endoscopy” in their full schedule. My special thanks also go to Lena An-dersson who assisted me tirelessly in implementing the wire-less pH-measuring technique.

The wonderful staff at 70 A2, led by Lotta Marquardt and guided by Gu-drun Berglund, for providing a pleasant environment in which to work and for their professional and dedicated care for our patients. I would also like to thank the rest of the staff at the surgical department.

Ann-Christin Linde at the Department of nuclear medicine for skillful and dedicated work with the scintigraphic studies.

All my climbing friends: Johan Sundström, Lars Pettersson, Johan Lundström, Jenny Kjellberg, Anders Tväråna, Magnus Crona and many more. For providing fantastic recuperation, fun climbing trips and sharing many instances when research or clinical work was definitely not the main priority.

All members of my family and extended family for help in many ways. My mother, Monica for raising me and always believing in my abilities and more recently for invaluable help with baby-sitting when our schedule is over-booked. My parents-in-law Karin and Bengt for much needed help with daily chores. My siblings Jesper, Mårten and Lotta for being role models and great companions.

61

Most of all I thank my wonderful wife and best friend Anna, whom I love deeply and who not only has endured me and supported me during all hard work of becoming a surgeon and doing research, but who has also developed a brilliant and impressive clinical and academic career of her own. And dur-ing all this, she has also given me two fantastic daughters, Ebba and Agnes, my everything, without whom the world would be infinitely less joyful and who give me love and inspiration in constantly new ways.

62

References

1. Gibbons A. Paleoanthropology. Food for thought. Science. Jun 15 2007;316(5831):1558-1560.

2. Verginelli F, Aru F, Battista P, Mariani-Costantini R. Nutrigenetics in the light of human evolution. J Nutrigenet Nutrigenomics. 2009;2(2):91-102.

3. Leonard WR, Snodgrass JJ, Robertson ML. Effects of brain evolution on human nutrition and metabolism. Annu Rev Nutr. 2007;27:311-327.

4. Ulijaszek SJ. Human eating behaviour in an evolutionary ecological context. Proc Nutr Soc. Nov 2002;61(4):517-526.

5. Cordain L, Eaton SB, Sebastian A, et al. Origins and evolution of the Western diet: health implications for the 21st century. Am J Clin Nutr. Feb 2005;81(2):341-354.

6. Deitel M. The obesity epidemic. Obes Surg. Apr 2006;16(4):377-378.

7. Ford ES, Mokdad AH. Epidemiology of obesity in the Western Hemisphere. J Clin Endocrinol Metab. Nov 2008;93(11 Suppl 1):S1-8.

8. Ogden CL, Yanovski SZ, Carroll MD, Flegal KM. The epidemiology of obesity. Gastroenterology. May 2007;132(6):2087-2102.

9. Sturm R. Increases in morbid obesity in the USA: 2000-2005. Public Health. Jul 2007;121(7):492-496.

10. Deitel M. Overweight and obesity worldwide now estimated to involve 1.7 billion people. Obes Surg. Jun 2003;13(3):329-330.

11. Hippocrates. Hippocrates I-VIII. Cambridge (MA): Loeb classical library, Harvard University Press; 1984-1995.

12. Bray GA, Greenway FL. Pharmacological treatment of the overweight patient. Pharmacol Rev. Jun 2007;59(2):151-184.

13. Scott HW, Jr., Dean R, Shull HJ, et al. Considerations in use of jejunoileal bypass in patients with morbid obesity. Ann Surg. Jun 1973;177(6):723-735.

14. Gomez CA. Gastroplasty in intractable obesity. Int J Obes. 1981;5(4):413-420.

15. Mason EE. Vertical banded gastroplasty for obesity. Arch Surg. May 1982;117(5):701-706.

16. Bo O, Modalsli O. Gastric banding, a surgical method of treating morbid obesity: preliminary report. Int J Obes. 1983;7(5):493-499.

63

17. Mason EE, Ito C. Gastric bypass in obesity. Surg Clin North Am. Dec 1967;47(6):1345-1351.

18. Wittgrove AC, Clark GW, Tremblay LJ. Laparoscopic Gastric Bypass, Roux-en-Y: Preliminary Report of Five Cases. Obes Surg. Nov 1994;4(4):353-357.

19. Hess DS, Hess DW. Biliopancreatic diversion with a duodenal switch. Obes Surg. Jun 1998;8(3):267-282.

20. Marceau P, Hould FS, Simard S, et al. Biliopancreatic diversion with duodenal switch. World J Surg. Sep 1998;22(9):947-954.

21. Scopinaro N, Gianetta E, Civalleri D, Bonalumi U, Bachi V. Bilio-pancreatic bypass for obesity: II. Initial experience in man. Br J Surg. Sep 1979;66(9):618-620.

22. Picot J, Jones J, Colquitt JL, et al. The clinical effectiveness and cost-effectiveness of bariatric (weight loss) surgery for obesity: a systematic review and economic evaluation. Health Technol Assess. Sep 2009;13(41):1-190, 215-357, iii-iv.

23. Buchwald H, Oien DM. Metabolic/bariatric surgery Worldwide 2008. Obes Surg. Dec 2009;19(12):1605-1611.

24. BMI classification. World Health Organization. http://www.who.int/bmi/index.jsp?introPage=intro_1.html.

25. Keil U, Kuulasmaa K. WHO MONICA Project: risk factors [published erratum appears in Int J Epidemiol 1990 Sep;19(3):following 775]. Int. J. Epidemiol. June 1, 1989 1989;18(suppl_1):S46-55.

26. 1983 metropolitan height and weight tables. Stat Bull Metrop Life Found. Jan-Jun 1983;64(1):3-9.

27. McFarlane SI, Banerji M, Sowers JR. Insulin resistance and cardiovascular disease. J Clin Endocrinol Metab. Feb 2001;86(2):713-718.

28. Jaffrin MY. Body composition determination by bioimpedance: an update. Curr Opin Clin Nutr Metab Care. Sep 2009;12(5):482-486.

29. McCrory MA, Gomez TD, Bernauer EM, Mole PA. Evaluation of a new air displacement plethysmograph for measuring human body composition. Med Sci Sports Exerc. Dec 1995;27(12):1686-1691.

30. Colleran GC, Moynagh MR, Tavernaraki K, Shelly MJ, Eustace SJ, Kavanagh EC. Whole-body magnetic resonance imaging: emerging applications. Semin Musculoskelet Radiol. Mar;14(1):57-67.

31. Olshansky SJ, Passaro DJ, Hershow RC, et al. A potential decline in life expectancy in the United States in the 21st century. N Engl J Med. Mar 17 2005;352(11):1138-1145.

32. Ogden CL CM, McDowell MA, Flegal KM. Obesity among adults in the United States-no change since 2003-2004. NCHS data brief no 1. Hyatsville, MD:National Center for Health Statistics. Vol 2007.

33. Eek F, Ostergren PO. Factors associated with BMI change over five years in a Swedish adult population. Results from the Scania Public Health Cohort Study. Scand J Public Health. Jul 2009;37(5):532-544.

64

34. Malin Kark ST, Finn Rasmussen. Övervikt och fetma i Stockholms län och Sverige- Tidstrender och samband med sociala förhållanden, psykiskt välbefinnande och självkänsla. In: Centrum för folkhälsa Sll, edStockholm: Edita Ljunglöfs; 2005.

35. Bostrom G, Eliasson M. Chapter 5.3: major public health problems - overweight and obesity. Scand J Public Health Suppl. Jun 2006;67:69-77.

36. Eckel RH, Barouch WW, Ershow AG. Report of the National Heart, Lung, and Blood Institute-National Institute of Diabetes and Digestive and Kidney Diseases Working Group on the pathophysiology of obesity-associated cardiovascular disease. Circulation. Jun 18 2002;105(24):2923-2928.

37. Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden associated with overweight and obesity. JAMA. Oct 27 1999;282(16):1523-1529.

38. Adams KF, Schatzkin A, Harris TB, et al. Overweight, obesity, and mortality in a large prospective cohort of persons 50 to 71 years old. N Engl J Med. Aug 24 2006;355(8):763-778.

39. Katzmarzyk PT, Janssen I, Ardern CI. Physical inactivity, excess adiposity and premature mortality. Obes Rev. Nov 2003;4(4):257-290.

40. Arnlov J, Ingelsson E, Sundstrom J, Lind L. Impact of body mass index and the metabolic syndrome on the risk of cardiovascular disease and death in middle-aged men. Circulation. Jan 19;121(2):230-236.

41. Fontaine KR, Redden DT, Wang C, Westfall AO, Allison DB. Years of life lost due to obesity. JAMA. Jan 8 2003;289(2):187-193.

42. Kirk EP, Klein S. Pathogenesis and pathophysiology of the cardiometabolic syndrome. J Clin Hypertens (Greenwich). Dec 2009;11(12):761-765.

43. Mathieu P, Poirier P, Pibarot P, Lemieux I, Despres JP. Visceral obesity: the link among inflammation, hypertension, and cardiovascular disease. Hypertension. Apr 2009;53(4):577-584.

44. Field AE, Coakley EH, Must A, et al. Impact of overweight on the risk of developing common chronic diseases during a 10-year period. Arch Intern Med. Jul 9 2001;161(13):1581-1586.

45. Oguma Y, Sesso HD, Paffenbarger RS, Jr., Lee IM. Weight change and risk of developing type 2 diabetes. Obes Res. May 2005;13(5):945-951.

46. Despres JP, Nadeau A, Tremblay A, et al. Role of deep abdominal fat in the association between regional adipose tissue distribution and glucose tolerance in obese women. Diabetes. Mar 1989;38(3):304-309.

47. Malhotra A, White DP. Obstructive sleep apnoea. Lancet. Jul 20 2002;360(9328):237-245.

65

48. Vgontzas AN, Bixler EO, Chrousos GP. Sleep apnea is a manifestation of the metabolic syndrome. Sleep Med Rev. Jun 2005;9(3):211-224.

49. AICR. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Washington DC: World Cancer Research Fund/American Institute for Cancer Research;2007.

50. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med. Apr 24 2003;348(17):1625-1638.

51. Wing RR, Phelan S. Long-term weight loss maintenance. Am J Clin Nutr. Jul 2005;82(1 Suppl):222S-225S.

52. Mann T, Tomiyama AJ, Westling E, Lew AM, Samuels B, Chatman J. Medicare's search for effective obesity treatments: diets are not the answer. Am Psychol. Apr 2007;62(3):220-233.

53. Colman E. Anorectics on trial: a half century of federal regulation of prescription appetite suppressants. Ann Intern Med. Sep 6 2005;143(5):380-385.

54. Sjostrom L, Rissanen A, Andersen T, et al. Randomised placebo-controlled trial of orlistat for weight loss and prevention of weight regain in obese patients. European Multicentre Orlistat Study Group. Lancet. Jul 18 1998;352(9123):167-172.

55. James WP, Astrup A, Finer N, et al. Effect of sibutramine on weight maintenance after weight loss: a randomised trial. STORM Study Group. Sibutramine Trial of Obesity Reduction and Maintenance. Lancet. Dec 23-30 2000;356(9248):2119-2125.

56. Colquitt J, Clegg A, Loveman E, Royle P, Sidhu MK. Surgery for morbid obesity. Cochrane Database Syst Rev. 2005(4):CD003641.

57. Payne JH, DeWind LT. Surgical treatment of obesity. Am J Surg. Aug 1969;118(2):141-147.

58. Ravitch MM, Brolin RE. The price of weight loss by jejunoileal shunt. Ann Surg. Sep 1979;190(3):382-391.

59. Salmon PA. The results of small intestine bypass operations for the treatment of obesity. Surg Gynecol Obstet. Jun 1971;132(6):965-979.

60. Scott HW, Jr., Dean RH, Shull HJ, et al. Further considerations in use of jejunoileal bypass in patients with morbid obesity. Bull Soc Int Chir. Sep-Dec 1974;33(5-6):378-387.

61. DeWind LT, Payne JH. Intestinal bypass surgery for morbid obesity. Long-term results. JAMA. Nov 15 1976;236(20):2298-2301.

62. Morino F, Toppino M, Fronda G, Tapparo A, Avagnina S. Weight Loss and Complications After Vertical Banded Gastroplasty. Obes Surg. Feb 1992;2(1):69-73.

63. Toppino M, Nigra II, Olivieri F, et al. Staple-line Disruptions in Vertical Banded Gastroplasty Related to Different Stapling Techniques. Obes Surg. Aug 1994;4(3):256-261.

66

64. Chevallier JM, Zinzindohoue F, Douard R, et al. Complications after laparoscopic adjustable gastric banding for morbid obesity: experience with 1,000 patients over 7 years. Obes Surg. Mar 2004;14(3):407-414.

65. Niville E, Dams A, Vlasselaers J. Lap-Band erosion: incidence and treatment. Obes Surg. Dec 2001;11(6):744-747.

66. Peterli R, Donadini A, Peters T, Ackermann C, Tondelli P. Re-operations following laparoscopic adjustable gastric banding. Obes Surg. Dec 2002;12(6):851-856.

67. Griffen WO, Jr., Young VL, Stevenson CC. A prospective comparison of gastric and jejunoileal bypass procedures for morbid obesity. Ann Surg. Oct 1977;186(4):500-509.

68. Torres JC, Oca CF, Garrison RN. Gastric bypass: Roux-en-Y gastrojejunostomy from the lesser curvature. South Med J. Oct 1983;76(10):1217-1221.

69. McCarthy HB, Rucker RD, Jr., Chan EK, et al. Gastritis after gastric bypass surgery. Surgery. Jul 1985;98(1):68-71.

70. Sundbom M, Gustavsson S. Randomized clinical trial of hand-assisted laparoscopic versus open Roux-en-Y gastric bypass for the treatment of morbid obesity. Br J Surg. Apr 2004;91(4):418-423.

71. Lujan JA, Frutos MD, Hernandez Q, et al. Laparoscopic versus open gastric bypass in the treatment of morbid obesity: a randomized prospective study. Ann Surg. Apr 2004;239(4):433-437.

72. Nguyen NT, Goldman C, Rosenquist CJ, et al. Laparoscopic versus open gastric bypass: a randomized study of outcomes, quality of life, and costs. Ann Surg. Sep 2001;234(3):279-289; discussion 289-291.

73. Westling A, Gustavsson S. Laparoscopic vs open Roux-en-Y gastric bypass: a prospective, randomized trial. Obes Surg. Jun 2001;11(3):284-292.

74. DeMeester TR, Fuchs KH, Ball CS, Albertucci M, Smyrk TC, Marcus JN. Experimental and clinical results with proximal end-to-end duodenojejunostomy for pathologic duodenogastric reflux. Ann Surg. Oct 1987;206(4):414-426.

75. Hess DS, Hess DW, Oakley RS. The biliopancreatic diversion with the duodenal switch: results beyond 10 years. Obes Surg. Mar 2005;15(3):408-416.

76. Prachand VN, Davee RT, Alverdy JC. Duodenal switch provides superior weight loss in the super-obese (BMI > or =50 kg/m2) compared with gastric bypass. Ann Surg. Oct 2006;244(4):611-619.

77. Strain GW, Gagner M, Inabnet WB, Dakin G, Pomp A. Comparison of effects of gastric bypass and biliopancreatic diversion with duodenal switch on weight loss and body composition 1-2 years after surgery. Surg Obes Relat Dis. Jan-Feb 2007;3(1):31-36.

78. Sovik TT, Taha O, Aasheim ET, et al. Randomized clinical trial of laparoscopic gastric bypass versus laparoscopic duodenal switch for superobesity. Br J Surg. Feb;97(2):160-166.

67

79. Deveney CW, MacCabee D, Marlink K, Welker K, Davis J, McConnell DB. Roux-en-Y divided gastric bypass results in the same weight loss as duodenal switch for morbid obesity. Am J Surg. May 2004;187(5):655-659.

80. Yurcisin BM, DeMaria EJ. Management of leak in the bariatric gastric bypass patient: reoperate, drain and feed distally. J Gastrointest Surg. Sep 2009;13(9):1564-1566.

81. Lee S, Carmody B, Wolfe L, et al. Effect of location and speed of diagnosis on anastomotic leak outcomes in 3828 gastric bypass cases. J Gastrointest Surg. Jun 2007;11(6):708-713.

82. Podnos YD, Jimenez JC, Wilson SE, Stevens CM, Nguyen NT. Complications after laparoscopic gastric bypass: a review of 3464 cases. Arch Surg. Sep 2003;138(9):957-961.

83. Bakhos C, Alkhoury F, Kyriakides T, Reinhold R, Nadzam G. Early postoperative hemorrhage after open and laparoscopic roux-en-y gastric bypass. Obes Surg. Feb 2009;19(2):153-157.

84. Nguyen NT, Longoria M, Chalifoux S, Wilson SE. Gastrointestinal hemorrhage after laparoscopic gastric bypass. Obes Surg. Nov-Dec 2004;14(10):1308-1312.

85. Omalu BI, Luckasevic T, Shakir AM, Rozin L, Wecht CH, Kuller LH. Postbariatric surgery deaths, which fall under the jurisdiction of the coroner. Am J Forensic Med Pathol. Sep 2004;25(3):237-242.

86. Melinek J, Livingston E, Cortina G, Fishbein MC. Autopsy findings following gastric bypass surgery for morbid obesity. Arch Pathol Lab Med. Sep 2002;126(9):1091-1095.

87. Parakh S, Soto E, Merola S. Diagnosis and management of internal hernias after laparoscopic gastric bypass. Obes Surg. Nov 2007;17(11):1498-1502.

88. Felsher J, Brodsky J, Brody F. Small bowel obstruction after laparoscopic Roux-en-Y gastric bypass. Surgery. Sep 2003;134(3):501-505.

89. Moize V, Geliebter A, Gluck ME, et al. Obese patients have inadequate protein intake related to protein intolerance up to 1 year following Roux-en-Y gastric bypass. Obes Surg. Feb 2003;13(1):23-28.

90. Aasheim ET, Bjorkman S, Sovik TT, et al. Vitamin status after bariatric surgery: a randomized study of gastric bypass and duodenal switch. Am J Clin Nutr. Jul 2009;90(1):15-22.

91. Alvarez-Leite JI. Nutrient deficiencies secondary to bariatric surgery. Curr Opin Clin Nutr Metab Care. Sep 2004;7(5):569-575.

92. Bloomberg RD, Fleishman A, Nalle JE, Herron DM, Kini S. Nutritional deficiencies following bariatric surgery: what have we learned? Obes Surg. Feb 2005;15(2):145-154.

93. Slater GH, Ren CJ, Siegel N, et al. Serum fat-soluble vitamin deficiency and abnormal calcium metabolism after malabsorptive bariatric surgery. J Gastrointest Surg. Jan 2004;8(1):48-55; discussion 54-45.

68

94. Shin JS, Chen KW, Lin XZ, Lin CY, Chang TT, Yang CC. Active, bleeding marginal ulcer of Billroth II gastric resection: a clinical experience of 18 patients. Am J Gastroenterol. Oct 1994;89(10):1831-1835.

95. Schirmer BD, Meyers WC, Hanks JB, Kortz WJ, Jones RS, Postlethwait RW. Marginal ulcer. A difficult surgical problem. Ann Surg. May 1982;195(5):653-661.

96. Spaulding L. The impact of small bowel resection on the incidence of stomal stenosis and marginal ulcer after gastric bypass. Obes Surg. Dec 1997;7(6):485-487; discussion 488.

97. Rasmussen JJ, Fuller W, Ali MR. Marginal ulceration after laparoscopic gastric bypass: an analysis of predisposing factors in 260 patients. Surg Endosc. Jul 2007;21(7):1090-1094.

98. Sapala JA, Wood MH, Sapala MA, Schuhknecht MP, Flake TM, Jr. The micropouch gastric bypass: technical considerations in primary and revisionary operations. Obes Surg. Feb 2001;11(1):3-17.

99. Jordan JH, Hocking MP, Rout WR, Woodward ER. Marginal ulcer following gastric bypass for morbid obesity. Am Surg. May 1991;57(5):286-288.

100. Patel RA, Brolin RE, Gandhi A. Revisional operations for marginal ulcer after Roux-en-Y gastric bypass. Surg Obes Relat Dis. Nov 6 2008.

101. Lublin M, McCoy M, Waldrep DJ. Perforating marginal ulcers after laparoscopic gastric bypass. Surg Endosc. Jan 2006;20(1):51-54.

102. Csendes A, Burgos AM, Altuve J, Bonacic S. Incidence of marginal ulcer 1 month and 1 to 2 years after gastric bypass: a prospective consecutive endoscopic evaluation of 442 patients with morbid obesity. Obes Surg. Feb 2009;19(2):135-138.

103. Capella JF, Capella RF. Gastro-gastric fistulas and marginal ulcers in gastric bypass procedures for weight reduction. Obes Surg. Feb 1999;9(1):22-27; discussion 28.

104. MacLean LD, Rhode BM, Nohr C, Katz S, McLean AP. Stomal ulcer after gastric bypass. J Am Coll Surg. Jul 1997;185(1):1-7.

105. Buchwald H, Estok R, Fahrbach K, Banel D, Sledge I. Trends in mortality in bariatric surgery: a systematic review and meta-analysis. Surgery. Oct 2007;142(4):621-632; discussion 632-625.

106. Kim WW, Gagner M, Kini S, et al. Laparoscopic vs. open biliopancreatic diversion with duodenal switch: a comparative study. J Gastrointest Surg. May-Jun 2003;7(4):552-557.

107. Sundbom M, Karlson BM. Low Mortality in Bariatric Surgery 1995 Through 2005 in Sweden, in Spite of a Shift to More Complex Procedures. Obes Surg. Oct 7 2008.

108. Schubert ML. Gastric secretion. Curr Opin Gastroenterol. Nov 2008;24(6):659-664.

109. Dimaline R, Varro A. Attack and defence in the gastric epithelium - a delicate balance. Exp Physiol. Jul 2007;92(4):591-601.

69

110. Mason EE, Munns JR, Kealey GP, et al. Effect of gastric bypass on gastric secretion. Am J Surg. Feb 1976;131(2):162-168.

111. Smith CD, Herkes SB, Behrns KE, Fairbanks VF, Kelly KA, Sarr MG. Gastric acid secretion and vitamin B12 absorption after vertical Roux-en-Y gastric bypass for morbid obesity. Ann Surg. Jul 1993;218(1):91-96.

112. Siilin H, Wanders A, Gustavsson S, Sundbom M. The proximal gastric pouch invariably contains acid-producing parietal cells in Roux-en-Y gastric bypass. Obes Surg. Jun-Jul 2005;15(6):771-777.

113. Siegel JA, Urbain JL, Adler LP, et al. Biphasic nature of gastric emptying. Gut. Jan 1988;29(1):85-89.

114. Ziessman HA, Fahey FH, Collen MJ. Biphasic solid and liquid gastric emptying in normal controls and diabetics using continuous acquisition in LAO view. Dig Dis Sci. May 1992;37(5):744-750.

115. Edin R. The vagal control of the pyloric motor function: a physiological and immunohistochemical study in cat and man. Acta Physiol Scand Suppl. 1980;485:1-30.

116. Rao SS, Lu C, Schulze-Delrieu K. Duodenum as a immediate brake to gastric outflow: a videofluoroscopic and manometric assessment. Gastroenterology. Mar 1996;110(3):740-747.

117. Ramkumar D, Schulze KS. The pylorus. Neurogastroenterol Motil. Jun 2005;17 Suppl 1:22-30.

118. Delzenne N, Blundell J, Brouns F, et al. Gastrointestinal targets of appetite regulation in humans. Obes Rev. Mar;11(3):234-250.

119. Naslund E, Gryback P, Backman L, et al. Distal small bowel hormones: correlation with fasting antroduodenal motility and gastric emptying. Dig Dis Sci. May 1998;43(5):945-952.

120. Lundberg JM, Tatemoto K, Terenius L, et al. Localization of peptide YY (PYY) in gastrointestinal endocrine cells and effects on intestinal blood flow and motility. Proc Natl Acad Sci U S A. Jul 1982;79(14):4471-4475.

121. Savage AP, Adrian TE, Carolan G, Chatterjee VK, Bloom SR. Effects of peptide YY (PYY) on mouth to caecum intestinal transit time and on the rate of gastric emptying in healthy volunteers. Gut. Feb 1987;28(2):166-170.

122. Imamura M. Effects of surgical manipulation of the intestine on peptide YY and its physiology. Peptides. Feb 2002;23(2):403-407.

123. Van Citters GW, Lin HC. The ileal brake: a fifteen-year progress report. Curr Gastroenterol Rep. Oct 1999;1(5):404-409.

124. Batterham RL, Cohen MA, Ellis SM, et al. Inhibition of food intake in obese subjects by peptide YY3-36. N Engl J Med. Sep 4 2003;349(10):941-948.

125. Vazquez Roque MI, Camilleri M, Stephens DA, et al. Gastric sensorimotor functions and hormone profile in normal weight, overweight, and obese people. Gastroenterology. Dec 2006;131(6):1717-1724.

70

126. Stock S, Leichner P, Wong AC, et al. Ghrelin, peptide YY, glucose-dependent insulinotropic polypeptide, and hunger responses to a mixed meal in anorexic, obese, and control female adolescents. J Clin Endocrinol Metab. Apr 2005;90(4):2161-2168.

127. Kim BJ, Carlson OD, Jang HJ, Elahi D, Berry C, Egan JM. Peptide YY is secreted after oral glucose administration in a gender-specific manner. J Clin Endocrinol Metab. Dec 2005;90(12):6665-6671.

128. Boggiano MM, Chandler PC, Oswald KD, et al. PYY3-36 as an anti-obesity drug target. Obes Rev. Nov 2005;6(4):307-322.

129. Naslund E, Gryback P, Hellstrom PM, et al. Gastrointestinal hormones and gastric emptying 20 years after jejunoileal bypass for massive obesity. Int J Obes Relat Metab Disord. May 1997;21(5):387-392.

130. Steinert RE, Poller B, Castelli MC, Drewe J, Beglinger C. Oral administration of glucagon-like peptide 1 or peptide YY 3-36 affects food intake in healthy male subjects. Am J Clin Nutr. Oct;92(4):810-817.

131. Holdstock C, Zethelius B, Sundbom M, Karlsson FA, Eden Engstrom B. Postprandial changes in gut regulatory peptides in gastric bypass patients. Int J Obes (Lond). Nov 2008;32(11):1640-1646.

132. Gryback P, Hermansson G, Lyrenas E, Beckman KW, Jacobsson H, Hellstrom PM. Nationwide standardisation and evaluation of scintigraphic gastric emptying: reference values and comparisons between subgroups in a multicentre trial. Eur J Nucl Med. Jun 2000;27(6):647-655.

133. Gryback P, Bajc M, Granerus G, et al. [Examination of gastric emptying with gamma camera. Clinically useful, uniform method now established]. Lakartidningen. Apr 12 2000;97(15):1811-1816.

134. Sapala JA, Wood MH, Sapala MA, Flake TM, Jr. Marginal ulcer after gastric bypass: a prospective 3-year study of 173 patients. Obes Surg. Oct 1998;8(5):505-516.

135. Owen DA. Normal histology of the stomach. Am J Surg Pathol. Jan 1986;10(1):48-61.

136. Chandrasoma PT, Der R, Ma Y, Dalton P, Taira M. Histology of the gastroesophageal junction: an autopsy study. Am J Surg Pathol. Mar 2000;24(3):402-409.

137. Lamb PJ, Griffin SM, Chandrashekar MV, Richardson DL, Karat D, Hayes N. Prospective study of routine contrast radiology after total gastrectomy. Br J Surg. Aug 2004;91(8):1015-1019.

138. Waldum HL, Qvigstad G, Fossmark R, Kleveland PM, Sandvik AK. Rebound acid hypersecretion from a physiological, pathophysiological and clinical viewpoint. Scand J Gastroenterol. Apr;45(4):389-394.

139. Chin EH, Hazzan D, Sarpel U, Herron DM. Multimedia article. Laparoscopic repair of a perforated marginal ulcer 2 years after gastric bypass. Surg Endosc. Nov 2007;21(11):2110.

71

140. Tuttle SG, Rufin F, Bettarello A. The physiology of heartburn. Ann Intern Med. Aug 1961;55:292-300.

141. Jamieson JR, Stein HJ, DeMeester TR, et al. Ambulatory 24-h esophageal pH monitoring: normal values, optimal thresholds, specificity, sensitivity, and reproducibility. Am J Gastroenterol. Sep 1992;87(9):1102-1111.

142. Piper DW, Fenton BH. pH stability and activity curves of pepsin with special reference to their clinical importance. Gut. Oct 1965;6(5):506-508.

143. Schriffin MJ. Production of experimental jejunal ulcer. Proc. Soc. exp. Biol. 1940;45:592-594.

144. Wilson JA, Romagnuolo J, Byrne TK, Morgan K, Wilson FA. Predictors of endoscopic findings after Roux-en-Y gastric bypass. Am J Gastroenterol. Oct 2006;101(10):2194-2199.

145. Marano BJ, Jr. Endoscopy after Roux-en-Y gastric bypass: a community hospital experience. Obes Surg. Mar 2005;15(3):342-345.

146. Sundbom M, Elphick DA, Mahida YR, et al. Alteration in human defensin-5 expression following gastric bypass surgery. J Clin Pathol. Sep 2007;60(9):1029-1034.

147. Abell TL, Camilleri M, Donohoe K, et al. Consensus recommendations for gastric emptying scintigraphy: a joint report of the American Neurogastroenterology and Motility Society and the Society of Nuclear Medicine. Am J Gastroenterol. Mar 2008;103(3):753-763.

148. Collins PJ, Horowitz M, Cook DJ, Harding PE, Shearman DJ. Gastric emptying in normal subjects--a reproducible technique using a single scintillation camera and computer system. Gut. Dec 1983;24(12):1117-1125.

149. Sanchez-Pernaute A, Rodriguez R, Rubio MA, et al. Gastric tube volume after duodenal switch and its correlation to short-term weight loss. Obes Surg. Sep 2007;17(9):1178-1182.

150. Bernstine H, Tzioni-Yehoshua R, Groshar D, et al. Gastric emptying is not affected by sleeve gastrectomy--scintigraphic evaluation of gastric emptying after sleeve gastrectomy without removal of the gastric antrum. Obes Surg. Mar 2009;19(3):293-298.

151. Braghetto I, Davanzo C, Korn O, et al. Scintigraphic Evaluation of Gastric Emptying in Obese Patients Submitted to Sleeve Gastrectomy Compared to Normal Subjects. Obes Surg. Aug 28 2009.

152. Maurer AH, Knight LC, Krevsky B. Proper definitions for lag phase in gastric emptying of solid foods. J Nucl Med. Mar 1992;33(3):466-467.

153. Mayer EA, Thomson JB, Jehn D, et al. Gastric emptying and sieving of solid food and pancreatic and biliary secretions after solid meals in patients with nonresective ulcer surgery. Gastroenterology. Dec 1984;87(6):1264-1271.

72

154. Adrian TE, Long RG, Fuessl HS, Bloom SR. Plasma peptide YY (PYY) in dumping syndrome. Dig Dis Sci. Dec 1985;30(12):1145-1148.

155. Johansson HE, Haenni A, Karlsson FA, et al. Bileopancreatic diversion with duodenal switch lowers both early and late phases of glucose, insulin and proinsulin responses after meal. Obes Surg. May;20(5):549-558.

156. Gould JC, Garren MJ, Boll V, Starling JR. Laparoscopic gastric bypass: risks vs. benefits up to two years following surgery in super-super obese patients. Surgery. Oct 2006;140(4):524-529; discussion 529-531.

157. Marceau S, Biron S, Lagace M, et al. Biliopancreatic Diversion, with Distal Gastrectomy, 250 cm and 50 cm Limbs: Long-term Results. Obes Surg. Aug 1995;5(3):302-307.

158. Parikh MS, Shen R, Weiner M, Siegel N, Ren CJ. Laparoscopic bariatric surgery in super-obese patients (BMI>50) is safe and effective: a review of 332 patients. Obes Surg. Jun-Jul 2005;15(6):858-863.

159. O'Rourke RW, Andrus J, Diggs BS, Scholz M, McConnell DB, Deveney CW. Perioperative morbidity associated with bariatric surgery: an academic center experience. Arch Surg. Mar 2006;141(3):262-268.

160. Gasteyger C, Suter M, Gaillard RC, Giusti V. Nutritional deficiencies after Roux-en-Y gastric bypass for morbid obesity often cannot be prevented by standard multivitamin supplementation. Am J Clin Nutr. May 2008;87(5):1128-1133.

161. Laurenius A, Taha O, Maleckas A, Lonroth H, Olbers T. Laparoscopic biliopancreatic diversion/duodenal switch or laparoscopic Roux-en-Y gastric bypass for super-obesity-weight loss versus side effects. Surg Obes Relat Dis. Jul-Aug;6(4):408-414.

162. Potoczna N, Harfmann S, Steffen R, Briggs R, Bieri N, Horber FF. Bowel habits after bariatric surgery. Obes Surg. Oct 2008;18(10):1287-1296.

163. Wasserberg N, Hamoui N, Petrone P, Crookes PF, Kaufman HS. Bowel habits after gastric bypass versus the duodenal switch operation. Obes Surg. Dec 2008;18(12):1563-1566.